{"slug":"brycewang-stanford-ai-model-benchmarking","source_name":"brycewang-stanford/ai-model-benchmarking","name":"Brycewang Stanford/AI Model Benchmarking","description":"Benchmark AI models across 60+ academic evaluation suites and metrics","version":1,"lift":{"pass_rate_delta_pts":13.64,"pass_rate_pct":95.5,"total_cases":22,"passed_cases":21,"tokens_delta_pct":100.3,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-29T00:13:16.743329+00:00"},"skill_score":0.9545,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":13.64,"with_pass_pct":95.5,"without_pass_pct":81.8,"tokens_delta_pct":100.3,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":21,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-29T00:13:16.743329+00:00","run_id":"35e95838-f40a-4ba5-b2b6-22be1208c634","latest_attempt":{"id":"35e95838-f40a-4ba5-b2b6-22be1208c634","runner_model":"gemini-3.6-flash","overall_verdict":"mixed","completed_at":"2026-08-29T00:13:16.743329+00:00","gate":null},"version_number":1,"is_latest_version":true,"gate":null}],"trust":{"skill_safety":"passed","safety_status":"clean","intent_verdict":"safe","content_status":"clean","indexable":true},"license":"NOASSERTION","install_count":0,"total_uses":0,"total_users":0,"manifest_hash":"d9c5e69fb1771b255f0978507a3b040b6cfa06dee5030126aa490590d36c745d","install_name":"brycewang-stanford-ai-model-benchmarking","version_id":"38ddea84-5561-49b6-97d0-db4c3d7bac83","body_sha256":"d9c5e69fb1771b255f0978507a3b040b6cfa06dee5030126aa490590d36c745d","render_hash":"efa8de396c9529d16d166d025d0f4c8dcebe7ba76f8dce9334f34312db70add8","raw_url":"https://app.decimal.ai/s/brycewang-stanford-ai-model-benchmarking/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/brycewang-stanford-ai-model-benchmarking"}