{"slug":"tetherto-vlm-benchmark","source_name":"tetherto/vlm-benchmark","name":"Tetherto/Vlm Benchmark","description":"Benchmark an optimization (PR or branch) on a Device Farm device via the vlm-benchmark framework — baseline vs optimized, quality-regression-aware.","version":1,"lift":{"pass_rate_delta_pts":41.67,"pass_rate_pct":75,"total_cases":24,"passed_cases":18,"tokens_delta_pct":391.6,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-29T12:07:23.466006+00:00"},"skill_score":0.7500000000000001,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":41.67,"with_pass_pct":75,"without_pass_pct":33.3,"tokens_delta_pct":391.6,"turns_delta_pct":0,"total_cases":24,"cases_aggregated":19,"verdict":"mixed","never_hurt":false,"completed_at":"2026-08-29T12:07:23.466006+00:00","run_id":"d38973ef-76cb-455f-a182-ed731a57e109","latest_attempt":{"id":"d38973ef-76cb-455f-a182-ed731a57e109","runner_model":"gemini-3.6-flash","overall_verdict":"mixed","completed_at":"2026-08-29T12:07:23.466006+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":"Apache-2.0","install_count":0,"total_uses":0,"total_users":0,"manifest_hash":"39a7929f4b3e060151818120dea8669e1f2e491355ca0276ad06106b1a827d21","install_name":"tetherto-vlm-benchmark","version_id":"f6e8d759-e0ad-43a8-89d5-a5b2c8e2d4cd","body_sha256":"39a7929f4b3e060151818120dea8669e1f2e491355ca0276ad06106b1a827d21","render_hash":"57194bb7ad4bdf2738b68343da1c2681c625afb6d781c9d93e8bcd83b4937958","raw_url":"https://app.decimal.ai/s/tetherto-vlm-benchmark/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/tetherto-vlm-benchmark"}