{"harness": {"slug": "appmap-navie-v2", "name": "Appmap Navie V2", "vendor": "AppMap", "description": "AppMap Navie is AppMap's AI coding agent with emphasis on application context and security-aware edits. v2 marks a public SWE-bench Verified submission generation.\n\nSee also `appmap-navie` for earlier submission label. `swe_bench_site` community scores.\n\nAppMap integrates with IDE workflows; public eval config may differ from enterprise deployments \u2014 rely on `source_url`.", "card_summary": "AppMap Navie v2 \u2014 security-aware coding agent on Verified.", "homepage_url": "https://appmap.io", "repo_url": "https://github.com/applandeo/appmap", "docs_url": null, "scaffold_code": null, "harbor_agent_name": null, "api_protocol": null, "status": "validated", "supports_rl": true, "score_count": 2, "best_success_rate": 47.2, "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/20241113_navie-2-gpt4o-sonnet.png", "aa_coding_index": null, "aa_mean_cost_usd": null, "aa_mean_total_tokens": null}, "scores": [{"success_rate": 47.2, "resolved_count": 236, "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://appmap.io/navie", "observed_at": "2024-11-06T00:00:00+00:00", "harness": {"slug": "appmap-navie-v2", "name": "Appmap Navie V2", "vendor": "AppMap", "repo_url": "https://github.com/applandeo/appmap", "openrouter_icon_url": null}, "model": {"slug": "gpt-4", "name": "GPT-4", "vendor": "openai"}, "benchmark": {"slug": "swe-bench-verified", "name": "SWE-bench Verified"}}, {"success_rate": 36.0, "resolved_count": 108, "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://appmap.io/navie", "observed_at": "2024-11-13T00:00:00+00:00", "harness": {"slug": "appmap-navie-v2", "name": "Appmap Navie V2", "vendor": "AppMap", "repo_url": "https://github.com/applandeo/appmap", "openrouter_icon_url": null}, "model": {"slug": "gpt-4", "name": "GPT-4", "vendor": "openai"}, "benchmark": {"slug": "swe-bench-lite", "name": "SWE-bench Lite"}}], "benchmarks": [{"slug": "swe-bench-lite", "name": "SWE-bench Lite", "category": "eval"}, {"slug": "swe-bench-verified", "name": "SWE-bench Verified", "category": "eval"}]}