deepseek-ai/DeepSeek-Math

DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

114 best points · 39 comments · 1 Hacker News post · 3,479 GitHub stars · View on GitHub

Hacker News discussions

• DeepSeekMath 7B achieved 51.7% on MATH benchmark (114 points, 39 comments, by mdp, 2024-02-06)