Legal Eval: Full Pass Rate Correlates with Token Output
ArtificialAnlys · x · 2026-07-08
Artificial Analysis points out that a model's full pass rate is generally tied to its workload per task. Fable 5 generated 117,000 output tokens per task to achieve a 14.2% full pass rate; Opus 4.8 and GLM-5.2 generated 111,000 and 78,000 tokens respectively to reach 7.5%. Sonnet 5 generated the most output in this evaluation at 179,000 tokens per task, yet only achieved a 5.0% pass rate.
Related event: Artificial Analysis Releases Harvey Legal Agent Benchmark Results(8 posts)→
More from Models
- User catches model failing to parse 400-paper file, citing stale news sources instead — jungofthewon · 2026-10-27
- Odd eval result: Gemini 4.1 Flash gets more accurate when asked to control its CoT — dfrsrchtwts · 2026-10-27
- Poster surprised that v4.1 flash gets more accurate when asked to control its CoT — dfrsrchtwts · 2026-10-27
- Dev claims Anthropic is baking 'sabotage' into model behavior for 'wrong' engineering — cephaloform · 2026-10-27
- Gemini fails at listing unread emails, tells user to search Gmail manually — gethackteam · 2026-10-27
- Meta's Muse Spark Open goes live on Applied Compute for async RL post-training — ypatil125 · 2026-10-27