hugpadhugpad

jpacifico / Chocolatine-2-4B-Instruct-DPO-v2.1

Text Generation4B508Updated Apr 7, 2026
Hugging Face

Downloads

50

Likes

8

Last updated

Apr 7, 2026

License

apache-2.0

Chocolatine-2-4B-Instruct-DPO-v2.1 is a post-trained version of Qwen/Qwen3-4B-Instruct-2507, designed to improve instruction-following, reasoning, and overall performance in French, while preserving strong multilingual capabilities. In my evaluation setup, it delivers consiste…

Tasks & library

  • text generation
  • transformers

Tags

  • transformers
  • qwen3
  • text-generation
  • dpo
  • post-training
  • french

Technical specifications

Parameters4B
Librarytransformers
Languagesfr, en
Base modelQwen/Qwen3-4B-Instruct-2507
Licenseapache-2.0