{"model": {"slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "vendor": "deepseek", "model_family": "deepseek-v4-pro", "parameter_count_b": null, "context_length": 1048576, "homepage_url": null, "hf_model_id": null, "openrouter_id": "deepseek/deepseek-v4-pro", "card_summary": "deepseek DeepSeek V4 Pro on SWE-bench Verified leaderboards.", "is_rl_checkpoint": false, "score_count": 6, "best_success_rate": 73.2, "input_price_per_million": 0.87, "output_price_per_million": 1.74}, "scores": [{"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": null, "openrouter_icon_url": null}, "model": {"slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "vendor": "deepseek"}, "benchmark": {"slug": "openhands-swe-bench", "name": "SWE-bench (OpenHands Index)"}}, {"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": null, "openrouter_icon_url": null}, "model": {"slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "vendor": "deepseek"}, "benchmark": {"slug": "swt-bench", "name": "SWT-Bench"}}, {"success_rate": 43.7, "resolved_count": null, "total_count": null, "cost_usd": 0.2, "input_tokens": null, "output_tokens": null, "latency_ms": null, "temperature": null, "max_turns": null, "context_budget_tokens": null, "harness_version": null, "trial_count": 60, "is_harnessrl_measured": false, "eval_protocol": "WildClawBench overall score (60 tasks)", "source": "wildclawbench_site", "source_url": "https://github.com/internlm/WildClawBench", "observed_at": null, "harness": {"slug": "openclaw", "name": "OpenClaw", "vendor": "OpenClaw", "repo_url": null, "openrouter_icon_url": null}, "model": {"slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "vendor": "deepseek"}, "benchmark": {"slug": "wildclawbench", "name": "WildClawBench"}}, {"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": null, "openrouter_icon_url": null}, "model": {"slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "vendor": "deepseek"}, "benchmark": {"slug": "swe-bench-multimodal", "name": "SWE-bench Multimodal (OpenHands Index)"}}, {"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": null, "openrouter_icon_url": null}, "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": 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": null, "openrouter_icon_url": null}, "model": {"slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "vendor": "deepseek"}, "benchmark": {"slug": "commit0", "name": "Commit0"}}], "harnesses": [{"slug": "openclaw", "name": "OpenClaw", "vendor": "OpenClaw", "repo_url": "https://github.com/openclaw/openclaw", "openrouter_icon_url": null}, {"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"}]}