423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-12-29 18:20:35 +08:00
Code Issues Packages Projects Releases Wiki Activity

[data] fix minicpmv plugin (#6801)

Browse Source 
* fix template name

* tiny fix

* support minicpm-o-2.6

* support inference of minicpmv

* update readme

* support dpo of minicpmv

* update init audio

* update init audio

* [model]fix image process in minicpmo

Former-commit-id: ab9bd068ef
This commit is contained in:
Zhangchi Feng
2025-02-04 21:20:15 +08:00
committed by GitHub
parent 822d5d362c
commit 85f22d01bf
3 changed files with 7 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/llamafactory/model/patcher.py
View File

@@ -125,7 +125,7 @@ def patch_config(
setattr(config, "use_cache", False) # qwen2 does not support use_cache when using flash attn
if getattr(config, "model_type", None) == "minicpmo":
setattr(config, "init_audio", False)
setattr(config, "init_audio", True)
setattr(config, "init_tts", False)
if "LlavaLlamaForCausalLM" in getattr(config, "architectures", []):