Commit Graph

218 Commits

Author SHA1 Message Date
Yaowei Zheng
75e6de5425 Merge commit from fork 2025-06-26 13:55:42 +08:00
Yaowei Zheng
336fd8bc09 [model] do not force load processor (#8457) 2025-06-25 19:43:00 +08:00
Yaowei Zheng
477bed6f89 [model] add kimi vl 2506 (#8432) 2025-06-23 17:56:48 +08:00
Vivek Iyer
4e0efdb48b [model] unsloth resume from checkpoint bug (#8423)
Co-authored-by: viyer <vivek_iyer2@apple.com>
2025-06-23 16:43:54 +08:00
Yaowei Zheng
3d05b82695 [model] fix vlm utils (#8388) 2025-06-17 01:08:49 +08:00
Yaowei Zheng
e62c9bc2bf [data] fix qwen2vl pos ids (#8387) 2025-06-17 00:48:54 +08:00
Yaowei Zheng
58f76eee69 [version] release v0.9.3 (#8386) 2025-06-16 19:21:32 +08:00
Kingsley
ae2a8100ba [model] support Mistral3.1 small 2503 (#8335) 2025-06-09 10:37:42 +08:00
Yaowei Zheng
5a21ca2448 [model] fix model generate (#8327) 2025-06-07 08:47:50 +08:00
Vivek Iyer
b441ecdbde [model] pushing FFT with unsloth (#8325)
Co-authored-by: viyer <vivek_iyer2@apple.com>
2025-06-07 08:20:58 +08:00
hoshi-hiyouga
abb581026f [deps] update to transformers 4.52 (#8125) 2025-05-21 05:16:18 +08:00
hoshi-hiyouga
8325087bb3 [model] switch to gptqmodel (#8108) 2025-05-19 22:25:40 +08:00
piamo
8dc195e4ad [model] update rope kwargs for yarn (#8101) 2025-05-19 20:07:54 +08:00
hoshi-hiyouga
d25c556714 [misc] update liger kernel patch (#7966) 2025-05-06 20:32:16 +02:00
hoshi-hiyouga
f7275e1ef5 [model] fix dsv3 leaf node (#7879) 2025-04-28 18:11:09 +08:00
zhaop-l
392a7879d9 [trainer] make projector trainable in freeze training (#7872)
Co-authored-by: hoshi-hiyouga <hiyouga@buaa.edu.cn>
2025-04-28 13:19:37 +08:00
Kingsley
3e2460bb38 fix attn patch for kimivl (#7867) 2025-04-27 23:12:28 +08:00
hoshi-hiyouga
95f92df771 [model] fix vit gradient checkpointing (#7830) 2025-04-23 22:48:48 +08:00
hoshi-hiyouga
d5f35f8b6c [model] fix moe zero3 (#7826) 2025-04-23 15:30:49 +08:00
Kingsley
000dbfe144 [data] fix internvl plugin (#7817) 2025-04-23 00:58:22 +08:00
Kingsley
19c49ef284 [model] add arch check for InternVL (#7803) 2025-04-22 16:38:05 +08:00
hoshi-hiyouga
acf641abc2 [data] improve mmplugin (#7795) 2025-04-22 01:25:33 +08:00
hoshi-hiyouga
cea9071ed1 [example] add bash usage (#7794) 2025-04-22 00:25:51 +08:00
flashJd
4d8f459ff6 [misc] fix new tokens adding (#7253)
Co-authored-by: hoshi-hiyouga <hiyouga@buaa.edu.cn>
2025-04-21 23:19:02 +08:00
ddddng
0313fbd8b0 [model] fix gemma3 export (#7786)
Co-authored-by: hoshi-hiyouga <hiyouga@buaa.edu.cn>
2025-04-21 23:07:11 +08:00
Sachin Beldona
1bae62b773 [misc] fix bug in constant (#7765)
Co-authored-by: Sachin Beldona <sbeldona@cs.cmu.edu>
2025-04-21 23:06:31 +08:00
hoshi-hiyouga
8208cbf1dc [trainer] fix pt loss (#7748)
* fix pt loss

* robust

* fix

* test
2025-04-17 03:15:35 +08:00
hoshi-hiyouga
a0818eae58 [breaking] bump transformers to 4.45.0 & improve ci (#7746)
* update ci

* fix

* fix

* fix

* fix

* fix
2025-04-17 02:36:48 +08:00
hoshi-hiyouga
06001ea2f0 [infer] set env for vllm ascend (#7745) 2025-04-17 01:08:55 +08:00
Kingsley
7a00670f70 [model] support intern-VL 2.5-3 series (#7258)
* add internvl and rebase

* fix for internvl2&3

* remove lines

* fix video_inputs & lint

* nit

* add constants

* remove lines

* fix

* fix error

* pass ci

* pass ci

* skip internvl & nit
2025-04-17 00:31:30 +08:00
Kingsley
d1b695cd9f [model] Support Kimi_VL thinking/instruct (#7719)
* add kimi_vl

* patch config

* check version

* Update mm_plugin.py

* Update mm_plugin.py

---------

Co-authored-by: hoshi-hiyouga <hiyouga@buaa.edu.cn>
2025-04-15 00:21:58 +08:00
hoshi-hiyouga
2b92e85cdd [misc] fix env vars (#7715) 2025-04-14 16:04:04 +08:00
hoshi-hiyouga
8f46aced51 [misc] upgrade cli (#7714) 2025-04-14 15:41:22 +08:00
Dain Kim
e60249d597 [bugfix] enable_gemma_liger_kernel (#7660)
- The `enable_liger_kernel` function for the Gemma model series was not executed due to the existing `if` statement in the code.
- Changed the line to an `elif` statement so that the `apply_liger_kernel` function is executed properly.

resolved: #7628
2025-04-10 11:27:30 +08:00
Kingsley
0935eff188 [data] Fix bugs of use_audio_in_video in Qwen2.5 Omni (#7638)
* cache _mm_inputs

* nit

* support for use_audio_in_video

* remove cache

* fix data

* Update mllm_video_audio_demo.json
2025-04-08 18:40:10 +08:00
hoshi-hiyouga
fb46193364 [misc] fix packing and eval plot (#7623) 2025-04-07 18:20:57 +08:00
hoshi-hiyouga
40fb24916f [model] add llama4 (#7611) 2025-04-06 13:42:31 +08:00
Kingsley
ac9ba80128 [data] fix qwen2.5 omni plugin (#7573)
* align key with qwen2vl

* nit && change scripts
2025-04-02 21:28:52 +08:00
hoshi-hiyouga
be0289292d [infer] vllm video/audio inference (#7566) 2025-04-02 02:27:04 +08:00
hoshi-hiyouga
37d783149d [model] fix kv cache (#7564) 2025-04-01 23:07:46 +08:00
Yu Shi Jie
69b0c1cf4f [model] fix use_cache patching for gemma3 multimodal (#7500) 2025-04-01 16:06:48 +08:00
Kingsley
1189aeb6c2 [model] add Qwen2.5-Omni model (#7537)
* preserve image_sizes

* preserve image_sizes

* init plugin

* support audio-text2text lora

* nit

* support image/video-text2text, audio-text2text

* remove args

* remove lines

* add docs && nit

* remove some comments

* fix && add merge part script

* add license
2025-03-31 20:39:35 +08:00
Xiaosu Zhu
d38c402f63 [misc] update liger-kernel's monkey patch (#7453)
* Update liger_kernel.py

* Update setup.py
2025-03-25 11:58:52 +08:00
AbdelKarim ELJANDOUBI
ce089ef8f6 [misc] enable liger kernel for gemma3 text and paligemma (#7466)
* add gemma3 text

* add paligemma (1,2 and 2 mix)
2025-03-25 09:27:43 +08:00
Kenny Lam
cad8bde6b1 [misc] enable liger kernel for gemma3 (#7462) 2025-03-24 19:09:59 +08:00
hoshi-hiyouga
e7ae755ab6 [data] gemma3 plugin pan and scan (#7294)
* gemma3 pan and scan

* add test case

* fix test
2025-03-13 23:29:23 +08:00
hoshi-hiyouga
1b1964714e [misc] update format (#7277) 2025-03-13 02:53:08 +08:00
hoshi-hiyouga
a54c859674 [model] support gemma3 (#7273) 2025-03-13 01:35:23 +08:00
hoshi-hiyouga
efa86e730c [misc] upgrade format to py39 (#7256) 2025-03-12 00:08:41 +08:00
hoshi-hiyouga
c6331546a9 [config] update args (#7231)
Former-commit-id: f71a901840811bf560df671ec63a146ff99140c6
2025-03-10 23:04:43 +08:00