{"schema_version":1,"total":1,"offset":0,"limit":100,"observations":[{"id":"public:56f66065c8216af200a5f9aa","benchmark_id":"vals-index-legal-research::2","subject":{"source_id":"alibaba/qwen3.6-plus","name":"alibaba/qwen3.6-plus","model_id":"qwen3.6-plus::default","variant":null,"harness":null},"value":14.904,"unit":"percent","basis":"measured","source":{"url":"https://www.vals.ai/benchmarks/vals_index","retrieved_at":"2026-09-18T13:00:43.826192+00:00","published_at":null,"file":"data/raw/benchmarks/daily-evidence/2026-09-18T12-58-20-378Z/558f8306d2b1cf1a1874.gz","sha256":"558f8306d2b1cf1a1874ebee5c907d17b0776ccc1f9e70fbb764af4470bb7bbb","locator":"astro_props; source row 45; alibaba/qwen3.6-plus; field accuracy"},"protocol":"Vals Index v2 (updated 2026-09-10), component legal_research; accuracy in percent, higher is better; each row is one model slug with its published compute/reasoning effort.; source row: {\"task\":\"legal_research\",\"accuracy\":14.904,\"stderr\":2.475,\"cost_per_test\":0.697997,\"latency\":658.954,\"compute_effort\":null,\"reasoning_effort\":null,\"max_output_tokens\":65536,\"provider\":\"Alibaba\"}","comparison_key":null,"join_note":"Reviewed identity map 2026-09-16: label names qwen3.6-plus without a setting; the catalog has exactly one configuration, the default (qwen3.6-plus::default)"}],"divergences":[],"cell":null,"coverage":{"model":null,"benchmark":{"total_models":844,"available":33,"unknown":811,"not_tested":0,"not_published":0,"source_unreachable":0,"contested":0,"measured":33,"self_reported":0,"observations":58,"unmatched_observations":25}},"collection":{"benchmark_id":"vals-index-legal-research::2","status":"collected","source_url":"https://www.vals.ai/benchmarks/vals_index","reason":"58 source results parsed; configurations remain separate; unmatched model identities are retained."},"coverage_note":"Denominators are catalog model configurations × versioned registry entries; capability_available excludes Efficiency (cost) boards. Basis counts count covered cells, not runs; measured and self_reported may overlap. Unmatched source identities remain in observations and never inflate catalog coverage."}