v2.5.1

English Version

New Features:

1. Support for RM for LLM and MLLM, as well as PPO for LLM.

New Models:

1. molmo series
2. mplug-owl3 1b/2b
3. llama3.1-nemotron-70b-instruct
4. deepseek-janus

中文版

新特性：

1. 支持LLM和MLLM的RM, 以及LLM的PPO.

新模型：

1. molmo系列
2. mplug-owl3 1b/2b
3. llama3.1-nemotron-70b-instruct
4. deepseek-janus

What's Changed

• support reward modeling and ppo by @hjh0119 in #2093
• fix rescale_image by @tastelikefeet in #2223
• fix deploy timeout by @Jintao-Huang in #2230
• Fix qwen2 vl batch size by @Jintao-Huang in #2239
• Fix ovis1.6 infer by @Jintao-Huang in #2242
• fix publish by @Jintao-Huang in #2245
• fix qwen2vl video args by @Jintao-Huang in #2251
• Update FAQ by @slin000111 in #2252
• Support molmo series vlm by @mi804 in #2260
• fix sft system by @Jintao-Huang in #2262
• support mplug3 1b/2b by @Jintao-Huang in #2271
• Fix deploy openai by @Jintao-Huang in #2278
• fix vllm ignore suffix by @Jintao-Huang in #2287
• fix lora_target_modules in PPO by @hjh0119 in #2274
• fix quant blocks by @Jintao-Huang in #2292
• Support Llama3.1-nemotron-70b-inst-hf by @DaozeZhang in #2299
• fix ppo citest by @hjh0119 in #2302
• support deepseek-janus by @Jintao-Huang in #2300
• update molmo by @Jintao-Huang in #2305

New Contributors

• @mi804 made their first contribution in #2260

Full Changelog: v2.5.0...v2.5.1