Sarvam-Prism scores 91.5% on 1,071 real decisions, 65ms each on one B200
itsOmSarraf_ · x · 2026-11-04
Atlan ran Sarvam-Prism through its Decision Bench: 1,071 real decisions across 36 public datasets, 91.5% accuracy on text rows — 12th of 22, level with Claude Sonnet 5 and Kev 27B, behind Amos, Perplexity Decider and Drex Max — at 65 ms per decision on a single B200. The accompanying trade-off charts show Decider V1.1 and Jev 1.1.3 leading on latency/cost while Drex Max tops accuracy.
More from Models
- LLM 'fixes' to a driving network work even with weights zeroed: model wrote its own controller — kevrussell · 2026-11-04
- Leaker hints: no "Fable 5.5" or "GPT-6.2 Astra" today — but Thursday is the date — daniel_mac8 · 2026-11-04
- Trajectory launches post-training platform: supply tasks, data and eval criteria, deploy trillion-parameter models — udmrzn · 2026-11-04
- Larger tokenizers suffer more multi-epoch regret, small vocabs survive repeats — SonglinYang4 · 2026-11-04
- Anime image forensics tool foidfinder detects character mashups and booru tags — AlpinDale · 2026-11-04
- Sonnet reportedly behaves with a different personality on Medium vs Max tiers — AaronBergman18 · 2026-11-04