The best Cohere model for science is North Mini Code (free), scoring 75.7% on GPQA Diamond — graduate-level physics, chemistry and biology. Command A (52.7%) is next.
AI models ranked by GPQA Diamond — graduate-level physics, chemistry and biology questions written to be un-Googleable, so the score reflects reasoning from knowledge rather than retrieval. The best large language models for scientific research and hard-science work.
The best Cohere model for science is North Mini Code (free), scoring 75.7% on GPQA Diamond — graduate-level physics, chemistry and biology. Command A (52.7%) is next.
Command A (52.7%) is the closest alternative on this metric. See the full ranking above for the tradeoffs.
modelgrep tracks 5 Cohere models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by North Mini Code (free). 2 of them qualify for this ranking.