sail/Qwen2.5-Math-7B-Oat-Zero

The sail/Qwen2.5-Math-7B-Oat-Zero model is a 7.6 billion parameter language model developed by sail, based on the Qwen2.5-Math-7B architecture. It is specifically fine-tuned using the minimalist R1-Zero recipe and Dr. DRPO algorithm on level 3-5 questions from the MATH dataset. This model is optimized for advanced mathematical reasoning and problem-solving tasks, demonstrating strong performance on widely used math benchmarks.

Warm
Public
7.6B
FP8
131072
License: apache-2.0
Hugging Face

No reviews yet. Be the first to review!