DeepSeek R1 Distill Qwen 7B

DeepSeek-R1-Distill-Qwen-7B is a 7 billion parameter dense language model distilled from DeepSeek-R1, utilizing reinforcement learning-enhanced reasoning data generated by DeepSeek's larger-scale models. The distillation process transfers advanced reasoning, mathematical, and coding capabilities into a smaller model architecture based on Qwen2.5-Math-7B. The model demonstrates strong performance across mathematics benchmarks (MATH-500 pass@1 of 92.8%), programming tasks (Codeforces rating of 1189), and general reasoning (GPQA Diamond pass@1 of 49.1%).