{"harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "description": "OpenHands SDK is the modern integration surface for the OpenHands project (All Hands AI). HarnessRL uses a thin adapter around the SDK agent loop: rollouts run inside Harbor sandboxes, and task success is scored by the Harbor verifier rather than ad-hoc string matching.\n\nThis is the default validated scaffold for many HarnessRL train/eval configs (`scaffold_code` `ohsdk`, Harbor `CustomOpenHandsSDK`). It supersedes the legacy `openhands` scaffold (`SCAFFOLD=oh`) for new work, though both share the upstream OpenHands repository.\n\nPublic scores appear under `swe_bench_site` (community submissions), `openhands_index` (OpenHands Index five-benchmark matrix), Artificial Analysis, and HarnessRL-measured rows when present.", "card_summary": "Primary OpenHands scaffold in HarnessRL \u2014 SDK agent loop with Harbor verifier reward.", "homepage_url": "https://github.com/All-Hands-AI/OpenHands", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "docs_url": "https://harness-rl.pages.dev/docs/architecture/agent-loop-workers", "scaffold_code": "ohsdk", "harbor_agent_name": "CustomOpenHandsSDK", "api_protocol": "openai", "status": "validated", "supports_rl": true, "score_count": 177, "best_success_rate": 95.8, "github_stars": 52300, "popularity_tier": "popular", "hrl_score": 59.0, "hrl_rank": 8, "catalog_token_volume": 565593629963, "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico", "leaderboard_icon_url": "https://www.swebench.com/img/logos/20241103_OpenHands-CodeAct-2.1-sonnet-20241022.png", "aa_coding_index": null, "aa_mean_cost_usd": null, "aa_mean_total_tokens": null}, "scores": [{"success_rate": 95.8, "resolved_count": null, "total_count": null, "cost_usd": 1.43, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/09f76346-1f26-4c2b-a2b9-d23108569bab", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "fable-5", "name": "Fable 5", "vendor": "Anthropic"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 91.9, "resolved_count": null, "total_count": null, "cost_usd": 1.47, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/e927fb1c-feb7-4d96-9e2c-f969cd6feccb", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "fable-5", "name": "Fable 5", "vendor": "Anthropic"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 86.1, "resolved_count": null, "total_count": null, "cost_usd": 0.74, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/268dd4d4-e587-449f-ba49-c2a3b6405683", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-5", "name": "GPT-5.5", "vendor": "openai"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 84.3, "resolved_count": null, "total_count": null, "cost_usd": 0.73, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/efde50d3-3151-48fc-aac2-c98a80d7406e", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-8-opus", "name": "Claude 4.8 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 84.2, "resolved_count": null, "total_count": null, "cost_usd": 7.91, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/1171bd32-73bf-4795-b4a3-40b42cf90fa3", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "fable-5", "name": "Fable 5", "vendor": "Anthropic"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 83.8, "resolved_count": null, "total_count": null, "cost_usd": 0.75, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/39ca1805-3f89-4219-8cdd-b6ca2360fb17", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-8-opus", "name": "Claude 4.8 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 83.4, "resolved_count": null, "total_count": null, "cost_usd": 0.92, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/6eec49a3-ecfb-4081-8282-caf1e3e1dda0", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-5", "name": "GPT-5.5", "vendor": "openai"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 82.4, "resolved_count": null, "total_count": null, "cost_usd": 0.61, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.13.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/9eaf668d-bfbe-4525-a0f3-d29cb6479a98", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-4", "name": "GPT-5.4", "vendor": "openai"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 81.8, "resolved_count": null, "total_count": null, "cost_usd": 1.8, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/3a131e0e-fdcd-4898-bfeb-7320496bbafd", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-1-pro", "name": "Gemini 3.1 Pro", "vendor": "Google"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 81.6, "resolved_count": null, "total_count": null, "cost_usd": 1.33, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/1444530f-c12e-43c7-a0b4-5847f0223a1e", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-7-opus", "name": "Claude 4.7 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 81.2, "resolved_count": null, "total_count": null, "cost_usd": 0.89, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/734845e8-b910-414f-a4b7-ad60d464131c", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-7-opus", "name": "Claude 4.7 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 81.2, "resolved_count": null, "total_count": null, "cost_usd": 2.63, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.24.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/0e76f325-8100-4491-8a2d-37fb1ba84d62", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-5-flash", "name": "Gemini 3.5 Flash", "vendor": "google"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 81.1, "resolved_count": null, "total_count": null, "cost_usd": 0.11, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/2203b457-5c29-4b00-be91-c0408f119244", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m3", "name": "MiniMax M3", "vendor": "minimax"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 80.8, "resolved_count": null, "total_count": null, "cost_usd": 0.82, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/96a4370b-a553-4c91-951b-79ae3a93979d", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-7-opus", "name": "Claude 4.7 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 80.0, "resolved_count": null, "total_count": null, "cost_usd": 0.44, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/09286fea-ded4-4f32-a8b1-97e3a5847b83", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-6-opus", "name": "Claude 4.6 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 78.8, "resolved_count": null, "total_count": null, "cost_usd": 1.17, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/d0d6404d-a144-4a7d-85db-8e1b50e071a8", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-8-opus", "name": "Claude 4.8 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 78.8, "resolved_count": null, "total_count": null, "cost_usd": 0.43, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/cbf7f857-21d0-4526-b830-25d4e88ab0ed", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-6-opus", "name": "Claude 4.6 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 78.6, "resolved_count": null, "total_count": null, "cost_usd": 1.53, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.24.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/60edc213-0c2b-4e3f-9c72-ba4bfce1cb32", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-5-flash", "name": "Gemini 3.5 Flash", "vendor": "google"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 78.5, "resolved_count": null, "total_count": null, "cost_usd": 1.38, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-5-opus", "name": "Claude 4.5 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 78.2, "resolved_count": null, "total_count": null, "cost_usd": 1.52, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/0875da38-90b0-4b49-9d66-0cbac446fd85", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-5", "name": "GPT-5.5", "vendor": "openai"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 76.8, "resolved_count": null, "total_count": null, "cost_usd": 0.77, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/77bb7715-dcff-4718-8e4f-611eae769107", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-6-opus", "name": "Claude 4.6 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 76.8, "resolved_count": null, "total_count": null, "cost_usd": 1.87, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/1e2fadc9-bfce-4a2d-80a2-da24b6e707ad", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-1-pro", "name": "Gemini 3.1 Pro", "vendor": "Google"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 76.6, "resolved_count": null, "total_count": null, "cost_usd": 1.82, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-5-opus", "name": "Claude 4.5 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 76.4, "resolved_count": null, "total_count": null, "cost_usd": 0.17, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/b56bbb3b-d00a-4bd8-93e0-554206f1dcba", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m3", "name": "MiniMax M3", "vendor": "minimax"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 76.2, "resolved_count": null, "total_count": null, "cost_usd": 1.03, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.24.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/88a0e23e-741b-4eba-a6a6-87056fb23905", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-5-flash", "name": "Gemini 3.5 Flash", "vendor": "google"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 75.6, "resolved_count": null, "total_count": null, "cost_usd": 0.63, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.13.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/5f6faa00-117b-4471-b635-5d830fcac6d5", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-4", "name": "GPT-5.4", "vendor": "openai"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 75.6, "resolved_count": null, "total_count": null, "cost_usd": 0.1795, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.14.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/a248d81f-d8d0-4823-9853-0378eed4dcc4", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-2-7", "name": "Minimax 2.7", "vendor": "MiniMax"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 75.1, "resolved_count": null, "total_count": null, "cost_usd": 1.62, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/c44c25a6-214d-4f87-aa3c-fdb8c3531d13", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-1-pro", "name": "Gemini 3.1 Pro", "vendor": "Google"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 75.0, "resolved_count": null, "total_count": null, "cost_usd": 1.54, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/6543c602-c721-48e6-a0fe-a1abb08fab4b", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-5-1", "name": "GLM 5 1", "vendor": "Zhipu"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 74.6, "resolved_count": null, "total_count": null, "cost_usd": 0.86, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-2", "name": "GPT-5.2", "vendor": "openai"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 74.6, "resolved_count": null, "total_count": null, "cost_usd": 0.67, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/e024c27b-0bc2-437c-8b36-f3695942dcef", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-6", "name": "KIMI K2.6", "vendor": "Moonshot"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 74.6, "resolved_count": null, "total_count": null, "cost_usd": 0.42, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-flash", "name": "Gemini 3 Flash", "vendor": "Google"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 74.5, "resolved_count": null, "total_count": null, "cost_usd": 0.42, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/f9cb5c66-98fb-48d9-87d6-08d7b16e66c3", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-6", "name": "KIMI K2.6", "vendor": "Moonshot"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 74.4, "resolved_count": null, "total_count": null, "cost_usd": 1.03, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/0e4ec892-21dc-424f-8802-b4db76c18123", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-6-sonnet", "name": "Claude 4.6 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 74.2, "resolved_count": null, "total_count": null, "cost_usd": 1.19, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-5-sonnet", "name": "Claude 4.5 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 74.2, "resolved_count": null, "total_count": null, "cost_usd": 1.5379, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/b9cb5405-1049-459f-8d4d-c256091a5549", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-6-plus", "name": "Qwen3.6 Plus", "vendor": "qwen"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 73.8, "resolved_count": null, "total_count": null, "cost_usd": 0.94, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-2-codex", "name": "GPT-5.2 codex", "vendor": "openai"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 73.4, "resolved_count": null, "total_count": null, "cost_usd": 1.06, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/58f92ff8-e886-40ab-a636-8337fc8c9bfe", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-5", "name": "GLM 5", "vendor": "z-ai"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 73.4, "resolved_count": null, "total_count": null, "cost_usd": 0.5596, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-4-7", "name": "GLM 4.7", "vendor": "z-ai"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 73.2, "resolved_count": null, "total_count": null, "cost_usd": 0.56, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-2", "name": "GPT-5.2", "vendor": "openai"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 73.2, "resolved_count": null, "total_count": null, "cost_usd": 0.0395, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.0.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/bec3b865-3408-4160-9d6a-c3830854c14e", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "vendor": "deepseek"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 72.7, "resolved_count": null, "total_count": null, "cost_usd": 0.87, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/b39f4bcc-7db3-435d-b7d1-517c269c16f2", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-5-sonnet", "name": "Claude 4.5 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 72.6, "resolved_count": null, "total_count": null, "cost_usd": 0.3066, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v0.38.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/5f0703c7-f946-414b-b3ba-f4ce77aad1b5", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m2-5", "name": "MiniMax M2.5", "vendor": "minimax"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 72.1, "resolved_count": null, "total_count": null, "cost_usd": 0.3377, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/b26a4b3d-0bc2-41d8-b8bf-09453cf2fb53", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-6-plus", "name": "Qwen3.6 Plus", "vendor": "qwen"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 71.8, "resolved_count": 359, "total_count": 500, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Verified public submission", "source": "swe_bench_site", "source_url": "https://all-hands.dev/", "observed_at": "2025-08-07T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5", "name": "GPT-5", "vendor": "openai"}, "benchmark": {"slug": "swe-bench-verified", "name": "SWE-bench Verified"}}, {"success_rate": 71.6, "resolved_count": null, "total_count": null, "cost_usd": 0.16, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "deepseek-v3-2-reasoner", "name": "DeepSeek V3.2 Reasoner", "vendor": "DeepSeek"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 70.9, "resolved_count": null, "total_count": null, "cost_usd": 0.55, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-2-codex", "name": "GPT-5.2 codex", "vendor": "openai"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 70.7, "resolved_count": null, "total_count": null, "cost_usd": 0.3, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-flash", "name": "Gemini 3 Flash", "vendor": "Google"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 70.6, "resolved_count": null, "total_count": null, "cost_usd": 4.39, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/53fd0c2a-c7bf-4efa-b74d-14258d81ed7a", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "fable-5", "name": "Fable 5", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 70.6, "resolved_count": null, "total_count": null, "cost_usd": 0.95, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-pro", "name": "Gemini 3 Pro", "vendor": "Google"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 70.4, "resolved_count": 352, "total_count": 500, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Verified public submission", "source": "swe_bench_site", "source_url": "https://all-hands.dev/", "observed_at": "2025-05-24T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-sonnet", "name": "Claude 4 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-verified", "name": "SWE-bench Verified"}}, {"success_rate": 70.4, "resolved_count": null, "total_count": null, "cost_usd": 0.47, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.13.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/90cebe80-49bb-40c1-b65b-78f5e40461f9", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-4", "name": "GPT-5.4", "vendor": "openai"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 70.4, "resolved_count": null, "total_count": null, "cost_usd": 0.33, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/9a4d9991-37b4-44a1-acff-2057c6525748", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-6", "name": "KIMI K2.6", "vendor": "Moonshot"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 70.2, "resolved_count": null, "total_count": null, "cost_usd": 0.76, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/edae3592-9be7-4da5-83c3-b6ebe10e1bf6", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-5-1", "name": "GLM 5 1", "vendor": "Zhipu"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 69.6, "resolved_count": 348, "total_count": 500, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Verified public submission", "source": "swe_bench_site", "source_url": "https://qwenlm.github.io/blog/qwen3-coder/", "observed_at": "2025-08-05T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-480b-a35b-instruct", "name": "Qwen3 Coder 480B A35B Instruct", "vendor": "Qwen"}, "benchmark": {"slug": "swe-bench-verified", "name": "SWE-bench Verified"}}, {"success_rate": 69.2, "resolved_count": null, "total_count": null, "cost_usd": 2.0, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-thinking", "name": "KIMI K2 Thinking", "vendor": "moonshotai"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 69.1, "resolved_count": null, "total_count": null, "cost_usd": 0.55, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-5-opus", "name": "Claude 4.5 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 69.1, "resolved_count": null, "total_count": null, "cost_usd": 0.1283, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.14.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/f42deebe-f174-41f7-96e8-3e2fbc8a61d7", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-2-7", "name": "Minimax 2.7", "vendor": "MiniMax"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 68.8, "resolved_count": null, "total_count": null, "cost_usd": 0.98, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-5-sonnet", "name": "Claude 4.5 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 68.8, "resolved_count": null, "total_count": null, "cost_usd": 0.4063, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-5", "name": "KIMI K2.5", "vendor": "moonshotai"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 68.8, "resolved_count": null, "total_count": null, "cost_usd": 0.7118, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m2-1", "name": "MiniMax M2.1", "vendor": "minimax"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 68.6, "resolved_count": null, "total_count": null, "cost_usd": 1.01, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-pro", "name": "Gemini 3 Pro", "vendor": "Google"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 68.1, "resolved_count": null, "total_count": null, "cost_usd": 0.2239, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v0.38.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/9bbf1e2a-6085-4a0a-b68f-357f417f4012", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m2-5", "name": "MiniMax M2.5", "vendor": "minimax"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 68.1, "resolved_count": null, "total_count": null, "cost_usd": 0.0354, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.0.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/4e693e26-5442-4d6a-9ee4-bdd26fc39213", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "vendor": "deepseek"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 67.3, "resolved_count": null, "total_count": null, "cost_usd": 0.47, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/0d35046a-cbee-473b-b316-d4f57ff7f9d2", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-5-1", "name": "GLM 5 1", "vendor": "Zhipu"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 67.0, "resolved_count": null, "total_count": null, "cost_usd": 0.66, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-2-codex", "name": "GPT-5.2 codex", "vendor": "openai"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 66.7, "resolved_count": null, "total_count": null, "cost_usd": 0.35, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/a3eb528f-9902-4f06-b083-4d5154a87e12", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m3", "name": "MiniMax M3", "vendor": "minimax"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 66.6, "resolved_count": null, "total_count": null, "cost_usd": 1.36, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/68c415d3-db32-418b-b0b4-2030edc2216e", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-next", "name": "Qwen3 Coder Next", "vendor": "qwen"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 65.5, "resolved_count": null, "total_count": null, "cost_usd": 0.48, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-2", "name": "GPT-5.2", "vendor": "openai"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 65.4, "resolved_count": 327, "total_count": 500, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Verified public submission", "source": "swe_bench_site", "source_url": "https://all-hands.dev/", "observed_at": "2025-07-16T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2", "name": "KIMI K2", "vendor": "moonshotai"}, "benchmark": {"slug": "swe-bench-verified", "name": "SWE-bench Verified"}}, {"success_rate": 63.6, "resolved_count": null, "total_count": null, "cost_usd": 0.3781, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-5", "name": "KIMI K2.5", "vendor": "moonshotai"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 62.5, "resolved_count": null, "total_count": null, "cost_usd": 12.49, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/95690e32-8d2e-4106-9a37-2f7d424a8494", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "fable-5", "name": "Fable 5", "vendor": "Anthropic"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 62.5, "resolved_count": null, "total_count": null, "cost_usd": 7.83, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/95f5d2b1-910c-403c-bb17-1b10a033470e", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-8-opus", "name": "Claude 4.8 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 62.4, "resolved_count": null, "total_count": null, "cost_usd": 1.26, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-480b", "name": "Qwen3 Coder 480B", "vendor": "Qwen"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 62.1, "resolved_count": null, "total_count": null, "cost_usd": 2.085, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/d5c3aa56-4cbd-4bac-9dfa-8d85cc25e994", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-6-plus", "name": "Qwen3.6 Plus", "vendor": "qwen"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 62.0, "resolved_count": null, "total_count": null, "cost_usd": 2.1639, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/828c5288-5e15-4660-9f22-c6bfe6e819e5", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-5-flash", "name": "Qwen3.5 Flash", "vendor": "Qwen"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 62.0, "resolved_count": null, "total_count": null, "cost_usd": 0.4663, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.14.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/054f0361-d26c-4466-82ae-042503b93f2e", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "nemotron-3-super", "name": "Nemotron 3 Super", "vendor": "NVIDIA"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 61.9, "resolved_count": null, "total_count": null, "cost_usd": 0.4246, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-5", "name": "KIMI K2.5", "vendor": "moonshotai"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 61.4, "resolved_count": null, "total_count": null, "cost_usd": 0.5441, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m2-1", "name": "MiniMax M2.1", "vendor": "minimax"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 60.8, "resolved_count": 304, "total_count": 500, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Verified public submission", "source": "swe_bench_site", "source_url": "https://all-hands.dev/", "observed_at": "2025-02-03T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "4x-scaled-2024-02-03", "name": "4x-scaled-2024-02-03", "vendor": "Research baseline"}, "benchmark": {"slug": "swe-bench-verified", "name": "SWE-bench Verified"}}, {"success_rate": 60.0, "resolved_count": null, "total_count": null, "cost_usd": 0.36, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/c37cd7c9-3839-402d-bdf6-a0c10c4c085b", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-5", "name": "GLM 5", "vendor": "z-ai"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 58.8, "resolved_count": null, "total_count": null, "cost_usd": 0.38, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-flash", "name": "Gemini 3 Flash", "vendor": "Google"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 56.8, "resolved_count": null, "total_count": null, "cost_usd": 0.69, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.16.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/8f5a30c8-5d86-4615-bc69-6d917e247944", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "trinity-large-thinking", "name": "Trinity Large Thinking", "vendor": "arcee-ai"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"success_rate": 56.2, "resolved_count": null, "total_count": null, "cost_usd": 5.69, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/825aa22d-16cb-49c1-8080-97a46db81099", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-7-opus", "name": "Claude 4.7 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 56.2, "resolved_count": null, "total_count": null, "cost_usd": 7.69, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/8a792401-ab45-48b6-8398-99a1a7534f69", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-6-opus", "name": "Claude 4.6 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 56.2, "resolved_count": null, "total_count": null, "cost_usd": 4.04, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.13.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/04c309cc-564d-465f-8e84-e0b114044609", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-4", "name": "GPT-5.4", "vendor": "openai"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 54.0, "resolved_count": null, "total_count": null, "cost_usd": 0.87, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/17b3fd34-7334-4c67-a510-cf2e094998c2", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-6-sonnet", "name": "Claude 4.6 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 53.9, "resolved_count": null, "total_count": null, "cost_usd": 0.15, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-4-7", "name": "GLM 4.7", "vendor": "z-ai"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 53.6, "resolved_count": null, "total_count": null, "cost_usd": 0.12, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "deepseek-v3-2-reasoner", "name": "DeepSeek V3.2 Reasoner", "vendor": "DeepSeek"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 53.0, "resolved_count": 265, "total_count": 500, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Verified public submission", "source": "swe_bench_site", "source_url": "https://docs.all-hands.dev/", "observed_at": "2024-10-29T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-3-5-sonnet-20241022", "name": "Claude 3.5 Sonnet (2024-10-22)", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-verified", "name": "SWE-bench Verified"}}, {"success_rate": 51.6, "resolved_count": 258, "total_count": 500, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Verified public submission", "source": "swe_bench_site", "source_url": "https://qwenlm.github.io/blog/qwen3-coder/", "observed_at": "2025-08-05T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-30b-a3b-instruct", "name": "Qwen3 Coder 30B A3B Instruct", "vendor": "qwen"}, "benchmark": {"slug": "swe-bench-verified", "name": "SWE-bench Verified"}}, {"success_rate": 50.9, "resolved_count": null, "total_count": null, "cost_usd": 0.13, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/04627eda-dae6-42d7-9533-669b5187640d", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-next", "name": "Qwen3 Coder Next", "vendor": "qwen"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 50.3, "resolved_count": null, "total_count": null, "cost_usd": 0.06, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "deepseek-v3-2-reasoner", "name": "DeepSeek V3.2 Reasoner", "vendor": "DeepSeek"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 50.0, "resolved_count": null, "total_count": null, "cost_usd": 1.81, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/472e594d-c161-4836-8a93-474c2e1e4168", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-8-opus", "name": "Claude 4.8 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 50.0, "resolved_count": null, "total_count": null, "cost_usd": 4.7, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/c343765e-1fdd-494f-96b7-de843c272a36", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-2", "name": "GPT-5.2", "vendor": "openai"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 50.0, "resolved_count": null, "total_count": null, "cost_usd": 6.48, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/a0e8861b-136f-4ff3-be1a-dcf2daa2428a", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-6-sonnet", "name": "Claude 4.6 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 49.7, "resolved_count": null, "total_count": null, "cost_usd": 0.3422, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/9a6fe5b9-dc75-41b4-a7e5-6c4817ae4c50", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-5-flash", "name": "Qwen3.5 Flash", "vendor": "Qwen"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 49.4, "resolved_count": null, "total_count": null, "cost_usd": 0.37, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-4-7", "name": "GLM 4.7", "vendor": "z-ai"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 48.5, "resolved_count": null, "total_count": null, "cost_usd": 2.83, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/cfe39593-1e41-44ba-bf2f-a29e64a51432", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-7-opus", "name": "Claude 4.7 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 47.9, "resolved_count": null, "total_count": null, "cost_usd": 0.0587, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v0.38.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/ac28bd0e-7131-47b4-ba01-e64d1c49df45", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m2-5", "name": "MiniMax M2.5", "vendor": "minimax"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 47.3, "resolved_count": null, "total_count": null, "cost_usd": 0.91, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/a7d29286-6810-4ca8-8a11-21d6ae33822b", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-5", "name": "GLM 5", "vendor": "z-ai"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 47.3, "resolved_count": null, "total_count": null, "cost_usd": 1.39, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-thinking", "name": "KIMI K2 Thinking", "vendor": "moonshotai"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 46.8, "resolved_count": 234, "total_count": 500, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Verified public submission", "source": "swe_bench_site", "source_url": "https://all-hands.dev/", "observed_at": "2025-05-20T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "devstral-small-2505", "name": "Devstral Small 2505", "vendor": "Mistral"}, "benchmark": {"slug": "swe-bench-verified", "name": "SWE-bench Verified"}}, {"success_rate": 45.7, "resolved_count": null, "total_count": null, "cost_usd": 1.1, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/a9302f07-92a9-48cb-9964-d91280be4d28", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-next", "name": "Qwen3 Coder Next", "vendor": "qwen"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 45.7, "resolved_count": null, "total_count": null, "cost_usd": 0.3333, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.14.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/01e58390-f947-4dcc-8d34-5e8c61bd3788", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "nemotron-3-super", "name": "Nemotron 3 Super", "vendor": "NVIDIA"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 44.2, "resolved_count": null, "total_count": null, "cost_usd": 0.5, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-pro", "name": "Gemini 3 Pro", "vendor": "Google"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 44.1, "resolved_count": null, "total_count": null, "cost_usd": 1.24, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/cca1e142-0f1b-45ef-95fc-fde8e4c9d3aa", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-1-pro", "name": "Gemini 3.1 Pro", "vendor": "Google"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 43.8, "resolved_count": null, "total_count": null, "cost_usd": 5.56, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/bab0c2b3-3c08-442d-98e8-7f34bfd76ebd", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-5", "name": "GPT-5.5", "vendor": "openai"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 43.8, "resolved_count": null, "total_count": null, "cost_usd": 5.5, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/57c6395a-4cd4-4346-a0c0-4b1209275523", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-2-codex", "name": "GPT-5.2 codex", "vendor": "openai"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 43.6, "resolved_count": null, "total_count": null, "cost_usd": 0.65, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-thinking", "name": "KIMI K2 Thinking", "vendor": "moonshotai"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 41.8, "resolved_count": null, "total_count": null, "cost_usd": 2.37, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/0fe00101-1f0a-4698-8a43-70fd665f4936", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-6-opus", "name": "Claude 4.6 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 41.67, "resolved_count": 125, "total_count": 300, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Lite public submission", "source": "swe_bench_site", "source_url": "https://docs.all-hands.dev/", "observed_at": "2024-10-25T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-3-5-sonnet-20241022", "name": "Claude 3.5 Sonnet (2024-10-22)", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-lite", "name": "SWE-bench Lite"}}, {"success_rate": 41.2, "resolved_count": null, "total_count": null, "cost_usd": 2.54, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-5-opus", "name": "Claude 4.5 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 41.2, "resolved_count": null, "total_count": null, "cost_usd": 6.92, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/26dfa179-9ab7-4b15-988d-b7ed2bf91328", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-5-1", "name": "GLM 5 1", "vendor": "Zhipu"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 41.2, "resolved_count": null, "total_count": null, "cost_usd": 0.64, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/bf318301-2c72-4fe2-a705-d104161d05c0", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-6", "name": "KIMI K2.6", "vendor": "Moonshot"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 40.6, "resolved_count": null, "total_count": null, "cost_usd": 0.2471, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m2-1", "name": "MiniMax M2.1", "vendor": "minimax"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 40.0, "resolved_count": null, "total_count": null, "cost_usd": 0.1206, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.14.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/83b65b31-9799-4019-844c-815151e4798e", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "nemotron-3-super", "name": "Nemotron 3 Super", "vendor": "NVIDIA"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 39.7, "resolved_count": null, "total_count": null, "cost_usd": 1.99, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.24.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/20d43172-1fef-4d83-af1b-01784887e7ca", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-5-flash", "name": "Gemini 3.5 Flash", "vendor": "google"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 38.3, "resolved_count": null, "total_count": null, "cost_usd": 1.7107, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/4ef023af-364e-4ba6-ac3d-19336bd4af18", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-5-flash", "name": "Qwen3.5 Flash", "vendor": "Qwen"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 38.2, "resolved_count": null, "total_count": null, "cost_usd": 2.81, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/55c33492-2ff8-4f59-ba0a-566bcdbb2cd4", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-5", "name": "GPT-5.5", "vendor": "openai"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 37.5, "resolved_count": null, "total_count": null, "cost_usd": 7.91, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.24.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/f73dcfa9-a6de-4b41-bcf6-abc1d2b0ba58", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-5-flash", "name": "Gemini 3.5 Flash", "vendor": "google"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 37.5, "resolved_count": null, "total_count": null, "cost_usd": 9.11, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/d15f9cba-5c46-4413-b71d-f7b5d15e6efb", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-5-opus", "name": "Claude 4.5 Opus", "vendor": "Anthropic"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 37.5, "resolved_count": null, "total_count": null, "cost_usd": 5.31, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/069f1865-a4b0-4a67-9451-3043c2ab138c", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-5-1", "name": "GLM 5 1", "vendor": "Zhipu"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 36.8, "resolved_count": null, "total_count": null, "cost_usd": 1.45, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.13.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/b8e55418-a747-461b-92b8-2981612e5650", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-4", "name": "GPT-5.4", "vendor": "openai"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 36.8, "resolved_count": null, "total_count": null, "cost_usd": 0.35, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/5cbf2dac-ffe8-47d0-9af4-db172c42e7b7", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m3", "name": "MiniMax M3", "vendor": "minimax"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 36.8, "resolved_count": null, "total_count": null, "cost_usd": 1.89, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-5-sonnet", "name": "Claude 4.5 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 36.8, "resolved_count": null, "total_count": null, "cost_usd": 1.46, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-pro", "name": "Gemini 3 Pro", "vendor": "Google"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 36.8, "resolved_count": null, "total_count": null, "cost_usd": 0.3746, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.0.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/926fc133-2a36-4c9c-b40b-353aa98057a9", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "vendor": "deepseek"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 35.9, "resolved_count": null, "total_count": null, "cost_usd": 2.97, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-2-codex", "name": "GPT-5.2 codex", "vendor": "openai"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 35.3, "resolved_count": null, "total_count": null, "cost_usd": 0.58, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/756a9904-f7f3-4c23-a663-8a7dd853d6dd", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-5", "name": "GLM 5", "vendor": "z-ai"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 34.9, "resolved_count": null, "total_count": null, "cost_usd": 0.97, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-480b", "name": "Qwen3 Coder 480B", "vendor": "Qwen"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 34.43, "resolved_count": 178, "total_count": 517, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Multimodal public submission", "source": "swe_bench_site", "source_url": "https://github.com/adityasoni9998/OpenHands-Versa", "observed_at": "2025-05-28T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "openhands-versa-claude4", "name": "Openhands Versa Claude4", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-multimodal-site", "name": "SWE-bench Multimodal (site)"}}, {"success_rate": 33.9, "resolved_count": null, "total_count": null, "cost_usd": 0.28, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-480b", "name": "Qwen3 Coder 480B", "vendor": "Qwen"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 33.3, "resolved_count": null, "total_count": null, "cost_usd": 0.8, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.16.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/5de44f24-5b10-4b75-a52d-05184f894a7e", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "trinity-large-thinking", "name": "Trinity Large Thinking", "vendor": "arcee-ai"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 32.8, "resolved_count": null, "total_count": null, "cost_usd": 1.6213, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-5", "name": "KIMI K2.5", "vendor": "moonshotai"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 32.7, "resolved_count": null, "total_count": null, "cost_usd": 0.35, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.16.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/74af64d2-66d2-4cb1-8d23-f981fafc88c6", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "trinity-large-thinking", "name": "Trinity Large Thinking", "vendor": "arcee-ai"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 32.4, "resolved_count": null, "total_count": null, "cost_usd": 2.31, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-thinking", "name": "KIMI K2 Thinking", "vendor": "moonshotai"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 31.33, "resolved_count": 162, "total_count": 517, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Multimodal public submission", "source": "swe_bench_site", "source_url": "https://github.com/adityasoni9998/OpenHands-Versa", "observed_at": "2025-05-09T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-3-7-sonnet", "name": "Claude 3.7 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-multimodal-site", "name": "SWE-bench Multimodal (site)"}}, {"success_rate": 31.2, "resolved_count": null, "total_count": null, "cost_usd": 2.23, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/76ea5306-b580-47c0-a9ad-67a1c995bfba", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-5", "name": "GLM 5", "vendor": "z-ai"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 30.9, "resolved_count": null, "total_count": null, "cost_usd": 2.77, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gpt-5-2", "name": "GPT-5.2", "vendor": "openai"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 30.9, "resolved_count": null, "total_count": null, "cost_usd": 2.3158, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/7c984e26-6009-4208-819f-a9ba57d5476f", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-6-plus", "name": "Qwen3.6 Plus", "vendor": "qwen"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 30.9, "resolved_count": null, "total_count": null, "cost_usd": 2.24, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/ed24d3b6-988c-4ef1-b709-b0f223173547", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-6-sonnet", "name": "Claude 4.6 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 30.9, "resolved_count": null, "total_count": null, "cost_usd": 1.52, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/8964f6e6-fcbd-4f26-a1ce-9fc657b8da43", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-next", "name": "Qwen3 Coder Next", "vendor": "qwen"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 29.38, "resolved_count": null, "total_count": null, "cost_usd": null, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "SWE-bench Test public submission", "source": "swe_bench_site", "source_url": "https://docs.all-hands.dev/", "observed_at": "2024-11-03T00:00:00+00:00", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-3-5-sonnet-20241022", "name": "Claude 3.5 Sonnet (2024-10-22)", "vendor": "Anthropic"}, "benchmark": {"slug": "swe-bench-test", "name": "SWE-bench Test"}}, {"success_rate": 27.9, "resolved_count": null, "total_count": null, "cost_usd": 0.19, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "deepseek-v3-2-reasoner", "name": "DeepSeek V3.2 Reasoner", "vendor": "DeepSeek"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 27.9, "resolved_count": null, "total_count": null, "cost_usd": 0.388, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.14.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/161dd621-b70d-4089-afcf-ea2e3d13e874", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-2-7", "name": "Minimax 2.7", "vendor": "MiniMax"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 27.9, "resolved_count": null, "total_count": null, "cost_usd": 2.6218, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/84c8dc14-d082-430d-b19c-e142de615385", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-5-flash", "name": "Qwen3.5 Flash", "vendor": "Qwen"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 25.5, "resolved_count": null, "total_count": null, "cost_usd": 0.2163, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.14.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/6839ba33-64fc-4b43-8ec2-e0c18f2f94bf", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-2-7", "name": "Minimax 2.7", "vendor": "MiniMax"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 25.0, "resolved_count": null, "total_count": null, "cost_usd": 1.18, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/3817b8ac-4a44-42f7-b6ce-08f747db0061", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-1-pro", "name": "Gemini 3.1 Pro", "vendor": "Google"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 25.0, "resolved_count": null, "total_count": null, "cost_usd": 0.62, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.18.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/9e4edd2a-981b-40fe-84dd-13c5fc992b42", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m3", "name": "MiniMax M3", "vendor": "minimax"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 25.0, "resolved_count": null, "total_count": null, "cost_usd": 1.52, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/65cee637-c137-4253-8fab-616dfbeda695", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-6", "name": "KIMI K2.6", "vendor": "Moonshot"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 25.0, "resolved_count": null, "total_count": null, "cost_usd": 7.4936, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/c5f31dca-5471-4ce3-830a-938de77baece", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-6-plus", "name": "Qwen3.6 Plus", "vendor": "qwen"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 25.0, "resolved_count": null, "total_count": null, "cost_usd": 3.18, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/bfba3513-7461-42da-9944-874848240c29", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-pro", "name": "Gemini 3 Pro", "vendor": "Google"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 25.0, "resolved_count": null, "total_count": null, "cost_usd": 0.57, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/df76145a-e319-468e-ae2e-07f71208496f", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "deepseek-v3-2-reasoner", "name": "DeepSeek V3.2 Reasoner", "vendor": "DeepSeek"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 25.0, "resolved_count": null, "total_count": null, "cost_usd": 0.5941, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v0.38.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/194e3e71-5874-45a5-96b8-220155ea1e24", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m2-5", "name": "MiniMax M2.5", "vendor": "minimax"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 25.0, "resolved_count": null, "total_count": null, "cost_usd": 1.04, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/0a1fbe3f-6704-42ad-8b8a-d598dbf493d8", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-next", "name": "Qwen3 Coder Next", "vendor": "qwen"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 25.0, "resolved_count": null, "total_count": null, "cost_usd": 0.74, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.16.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/f327bc7a-c85b-4a89-b0c9-2da1532bcec4", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "trinity-large-thinking", "name": "Trinity Large Thinking", "vendor": "arcee-ai"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 23.5, "resolved_count": null, "total_count": null, "cost_usd": 2.09, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-480b", "name": "Qwen3 Coder 480B", "vendor": "Qwen"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 22.1, "resolved_count": null, "total_count": null, "cost_usd": 0.8, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-flash", "name": "Gemini 3 Flash", "vendor": "Google"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 22.1, "resolved_count": null, "total_count": null, "cost_usd": 0.66, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-4-7", "name": "GLM 4.7", "vendor": "z-ai"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 20.6, "resolved_count": null, "total_count": null, "cost_usd": 0.7679, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.14.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/44cc701e-6ecf-4c70-9cca-4246cbd402f4", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "nemotron-3-super", "name": "Nemotron 3 Super", "vendor": "NVIDIA"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 18.8, "resolved_count": null, "total_count": null, "cost_usd": 1.26, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/7f44e8ae-607d-46e7-8f5d-b43abacf8c1b", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-5", "name": "KIMI K2.5", "vendor": "moonshotai"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 18.8, "resolved_count": null, "total_count": null, "cost_usd": 2.62, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/3501752b-0081-4611-9807-335281599b80", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "gemini-3-flash", "name": "Gemini 3 Flash", "vendor": "Google"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 18.8, "resolved_count": null, "total_count": null, "cost_usd": 0.8926, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.14.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/081a0b7f-fef2-4dad-a267-49f90b2fc866", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-2-7", "name": "Minimax 2.7", "vendor": "MiniMax"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 18.8, "resolved_count": null, "total_count": null, "cost_usd": 0.38, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/6b9df675-af26-4bd6-89b5-d18b38782a92", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m2-1", "name": "MiniMax M2.1", "vendor": "minimax"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 16.2, "resolved_count": null, "total_count": null, "cost_usd": 1.4043, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m2-1", "name": "MiniMax M2.1", "vendor": "minimax"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"success_rate": 13.3, "resolved_count": null, "total_count": null, "cost_usd": 0.41, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/452a6943-1a58-4ef5-a2c7-ab2328fa4501", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-6-sonnet", "name": "Claude 4.6 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 12.7, "resolved_count": null, "total_count": null, "cost_usd": 0.1594, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.0.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/ae011071-8444-4be2-87f3-99fde238eff1", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "vendor": "deepseek"}, "benchmark": {"slug": "gaia", "name": "GAIA"}}, {"success_rate": 12.5, "resolved_count": null, "total_count": null, "cost_usd": 3.23, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/17fc5d57-6aae-4dda-8f56-f6275df7db8e", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "claude-4-5-sonnet", "name": "Claude 4.5 Sonnet", "vendor": "Anthropic"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 12.5, "resolved_count": null, "total_count": null, "cost_usd": 0.49, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v0.38.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/c06b32df-bf9e-4d10-9e81-93f2c01955a5", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "minimax-m2-5", "name": "MiniMax M2.5", "vendor": "minimax"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 12.5, "resolved_count": null, "total_count": null, "cost_usd": 1.64, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/84ad9e18-dc52-4085-bc7f-54205df4fae4", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "glm-4-7", "name": "GLM 4.7", "vendor": "z-ai"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 12.5, "resolved_count": null, "total_count": null, "cost_usd": 0.54, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/411cb079-1f09-4779-96b9-4c658df2c0be", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "kimi-k2-thinking", "name": "KIMI K2 Thinking", "vendor": "moonshotai"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 12.5, "resolved_count": null, "total_count": null, "cost_usd": 0.1577, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.0.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/413fa09a-4b62-451f-9139-9d140a4d108c", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "vendor": "deepseek"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 12.5, "resolved_count": null, "total_count": null, "cost_usd": 4.3309, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.11.5", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/2416b7ce-7841-4d7b-b8fb-83ad5c992826", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-5-flash", "name": "Qwen3.5 Flash", "vendor": "Qwen"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 12.5, "resolved_count": null, "total_count": null, "cost_usd": 1.2153, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.14.0", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/ba725a00-2947-484f-9dad-951071d34c3d", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "nemotron-3-super", "name": "Nemotron 3 Super", "vendor": "NVIDIA"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 12.5, "resolved_count": null, "total_count": null, "cost_usd": 1.2, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.16.1", "source": "openhands_index", "source_url": "https://laminar.sh/shared/evals/8d97d9a6-f993-44f6-aacd-a6977c811bd3", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "trinity-large-thinking", "name": "Trinity Large Thinking", "vendor": "arcee-ai"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}, {"success_rate": 0.0, "resolved_count": null, "total_count": null, "cost_usd": 0.01, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": null, "is_harnessrl_measured": false, "eval_protocol": "OpenHands SDK v1.8.3", "source": "openhands_index", "source_url": "https://index.openhands.dev/", "observed_at": "2026-07-31T11:38:58.577710", "harness": {"slug": "openhands-sdk", "name": "OpenHands SDK", "vendor": "All Hands AI", "repo_url": "https://github.com/All-Hands-AI/OpenHands", "openrouter_icon_url": "https://www.all-hands.dev/favicon.ico"}, "model": {"slug": "qwen3-coder-480b", "name": "Qwen3 Coder 480B", "vendor": "Qwen"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}], "benchmarks": [{"slug": "commit0", "name": "Commit0", "category": "eval"}, {"slug": "gaia", "name": "GAIA", "category": "eval"}, {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)", "category": "eval"}, {"slug": "swe-bench-lite", "name": "SWE-bench Lite", "category": "eval"}, {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)", "category": "eval"}, {"slug": "swe-bench-multimodal-site", "name": "SWE-bench Multimodal (site)", "category": "eval"}, {"slug": "swe-bench-test", "name": "SWE-bench Test", "category": "eval"}, {"slug": "swe-bench-verified", "name": "SWE-bench Verified", "category": "eval"}, {"slug": "swt-bench", "name": "SWT-Bench", "category": "eval"}]}