{"harness": {"slug": "sima", "name": "SIMA", "vendor": "Swe Bench", "description": "SIMA is tracked in this catalog as a software-engineering agent harness. Public benchmark rows use this slug on leaderboards such as SWE-bench Verified or Artificial Analysis coding-agent evals.\n\nCheck each score row's `source` and `source_url` for the authoritative snapshot. Harness-level metadata here does not invent scores \u2014 only documents what the harness is and where published results come from.", "card_summary": "Swe Bench SIMA coding agent harness.", "homepage_url": "https://github.com/swe-bench/experiments/tree/main/evaluation/lite/20240706_sima_gpt4o", "repo_url": "https://github.com/swe-bench/experiments/tree/main/evaluation/lite/20240706_sima_gpt4o", "docs_url": null, "scaffold_code": null, "harbor_agent_name": null, "api_protocol": null, "status": "validated", "supports_rl": true, "score_count": 1, "best_success_rate": 27.67, "github_stars": 280, "popularity_tier": null, "hrl_score": 29.0, "hrl_rank": 40, "catalog_token_volume": null, "openrouter_icon_url": null, "leaderboard_icon_url": null, "aa_coding_index": null, "aa_mean_cost_usd": null, "aa_mean_total_tokens": null}, "scores": [{"success_rate": 27.67, "resolved_count": 83, "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://github.com/swe-bench/experiments/tree/main/evaluation/lite/20240706_sima_gpt4o", "observed_at": "2024-07-06T00:00:00+00:00", "harness": {"slug": "sima", "name": "SIMA", "vendor": "Swe Bench", "repo_url": "https://github.com/swe-bench/experiments/tree/main/evaluation/lite/20240706_sima_gpt4o", "openrouter_icon_url": null}, "model": {"slug": "gpt-4o-20240513", "name": "GPT-4o (2024-05-13)", "vendor": "OpenAI"}, "benchmark": {"slug": "swe-bench-lite", "name": "SWE-bench Lite"}}], "benchmarks": [{"slug": "swe-bench-lite", "name": "SWE-bench Lite", "category": "eval"}]}