{"harness": {"slug": "google-jules", "name": "Google Jules", "vendor": "Google", "description": "Google Jules is Google's agent product for asynchronous coding tasks, with public SWE-bench Verified submissions under the Jules harness name. Scores reflect Google's agent tooling, safety, and retrieval paired with disclosed Gemini or other Google models.\n\nProduct harness \u2014 not minimal bash-only baseline. `swe_bench_site` on `swe-bench-verified`. For Google's terminal CLI scaffold see `gemini-cli`; Jules rows are product-specific.", "card_summary": "Google Jules coding agent \u2014 SWE-bench Verified submission harness.", "homepage_url": "https://jules.google", "repo_url": null, "docs_url": null, "scaffold_code": null, "harbor_agent_name": null, "api_protocol": null, "status": "validated", "supports_rl": true, "score_count": 1, "best_success_rate": 52.2, "github_stars": null, "popularity_tier": null, "hrl_score": null, "hrl_rank": null, "catalog_token_volume": null, "openrouter_icon_url": null, "leaderboard_icon_url": "https://www.swebench.com/img/logos/20241212_google_jules_gemini_2.0_flash_experimental.png", "aa_coding_index": null, "aa_mean_cost_usd": null, "aa_mean_total_tokens": null}, "scores": [{"success_rate": 52.2, "resolved_count": 261, "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://developers.googleblog.com/en/the-next-chapter-of-the-gemini-era-for-developers/", "observed_at": "2024-12-12T00:00:00+00:00", "harness": {"slug": "google-jules", "name": "Google Jules", "vendor": "Google", "repo_url": null, "openrouter_icon_url": null}, "model": {"slug": "gemini-2-0-flash-20241212", "name": "Gemini 2.0 Flash (2024-12-12)", "vendor": "Google"}, "benchmark": {"slug": "swe-bench-verified", "name": "SWE-bench Verified"}}], "benchmarks": [{"slug": "swe-bench-verified", "name": "SWE-bench Verified", "category": "eval"}]}