Jackrong/gpt-oss-120b-Distill-Llama3.1-8B-v2

Soren's gpt-oss-120b-Distill-Llama3.1-8B-v2 is an 8 billion parameter Llama 3.1-based model, specifically engineered to distill advanced reasoning capabilities, including Chain-of-Thought (CoT), from larger teacher models. It utilizes a two-stage training process involving Supervised Fine-Tuning (SFT) and Reinforcement Learning (GRPO) to enhance logical and mathematical problem-solving. This model excels at generating structured thought processes and accurate solutions, particularly in mathematical reasoning, and supports both English and Chinese.