mychen76/ qwen2.5-3b-think-r1

257 1 year ago

A regular model convert into Reasoning/Think Model fine-tuned using DeepSeek GRPO algorithm without using distilled data from R1.

tools

2 models

qwen2.5-3b-think-r1:latest
0937053a5fde • 3.3GB • 32K context window •
Text input • 1 year ago
qwen2.5-3b-think-r1:q8 latest
0937053a5fde • 3.3GB • 32K context window •
Text input • 1 year ago