Needs a toolchainMIT Tested
DeepSeek-R1-Distill-Qwen-7B
In one line：DeepSeek-R1-Distill-Qwen-7B 是一个把推理能力蒸馏到 7B 的模型，回答前先输出思考过程。
把"先想再答"的能力压进 7B，本地也能做推理题。
Official source
We host and mirror nothing — the official page is the source of truth.
In one line：DeepSeek-R1-Distill-Qwen-7B 是一个把推理能力蒸馏到 7B 的模型，回答前先输出思考过程。
把"先想再答"的能力压进 7B，本地也能做推理题。
Official source
We host and mirror nothing — the official page is the source of truth.