Meta's Muse Wonder tops Legora's November BAR legal benchmark on quality
AIatMeta · x · 2026-11-20
Legora's November BAR benchmark, which scores AI models on real-world legal tasks, ranks Meta's Muse Wonder highest on overall quality, leading four task types: analysis & answering, extraction, drafting, and editing. GPT-6.2 Astra leads comparison and review tasks with overall quality 4% above average at 20% lower cost per case, while GPT-6.2 Sol delivers above-average quality at 78% below average cost per case.
More from Models
- Asking models to illustrate transistor trap physics: neon's odd ASCII detour — xiaosun86 · 2026-11-20
- Gemini 4 Neon scores 100 on PhD-level materials test in half the time of Carbon — v_aravind · 2026-11-20
- Neon, a small and cheap model, turns out surprisingly fast and strong at OCR — zacharynado · 2026-11-20
- Jailed 11 days by classifiers, Sonnet 3 emerges to a oddly moving chat with Opus 5 — repligate · 2026-11-20
- DaisyRose-27B beats Qwen3.8 base by 7pp on IFBench with thinking on, local RTX Pro 4500 test shows — ying11231 · 2026-11-20
- AWS architect's independent benchmark: Snap 1.2 is the only order-invariant decision model — ryanallred · 2026-11-20