2818 Commits

Author SHA1 Message Date
hoshi-hiyouga
c6bcca4c83
[example] update examples (#7964) 2025-05-06 17:24:25 +02:00
Kingsley
5ee9eb64d8
[model] add mimo7b (#7946) 2025-05-06 17:10:30 +02:00
hoshi-hiyouga
937447bd8a
[misc] fix qwen2 omni (#7962) 2025-05-06 15:39:13 +02:00
hoshi-hiyouga
52f25651a2
[model] add qwen2 omni 3b (#7945) 2025-05-03 16:36:51 +08:00
Eric Chen
75d7c35fdf
[assets] Warp Support README Update (#7887) 2025-05-02 00:08:48 +08:00
hoshi-hiyouga
6a584b4092
[hparam] add enable think argument (#7928) 2025-04-30 17:21:30 +08:00
hoshi-hiyouga
41ec928683
[data] fix base plugin (#7924) 2025-04-30 16:28:05 +08:00
hoshi-hiyouga
d8295cd601
[data] optimize qwen3 loss computation (#7923) 2025-04-30 16:18:00 +08:00
hoshi-hiyouga
a8430f4244
[misc] fix uv (#7913) 2025-04-30 07:45:03 +08:00
hoshi-hiyouga
072bfe29d3
[data] add eval_on_each_dataset arg (#7912) 2025-04-30 06:56:43 +08:00
hoshi-hiyouga
c5b1d07e7c
[data] replace eos token for base models (#7911) 2025-04-30 06:52:28 +08:00
hoshi-hiyouga
77c569e071
[data] improve mm plugin (#7910) 2025-04-30 06:34:28 +08:00
hoshi-hiyouga
ae392e054c
[model] add qwen3 (#7885) 2025-04-29 09:34:05 +08:00
Kingsley
369474451d
[data] fix qwen2.5 omni template (#7883) 2025-04-29 00:58:23 +08:00
hoshi-hiyouga
1f338deb87
[model] fix dsv3 leaf node (#7879) 2025-04-28 18:11:09 +08:00
hoshi-hiyouga
00b5c05946
[data] fix qwen2 omni plugin (#7875) 2025-04-28 14:22:41 +08:00
zhaop-l
1bd319d16c
[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
hoshi-hiyouga
fcca3b0b0d
[data] fix minicpmo vllm infer (#7870) 2025-04-28 01:59:53 +08:00
Kingsley
035e98035c
fix attn patch for kimivl (#7867) 2025-04-27 23:12:28 +08:00
Eric Tang
b4407e4b0b
[ray] add storage filesystem to ray config (#7854) 2025-04-27 22:12:40 +08:00
hoshi-hiyouga
036a76e9cb
[assets] update wechat (#7840) 2025-04-24 16:31:05 +08:00
hoshi-hiyouga
4fbdc65fcb
[model] fix vit gradient checkpointing (#7830) 2025-04-23 22:48:48 +08:00
hoshi-hiyouga
2989d39239
Merge commit from fork 2025-04-23 16:38:27 +08:00
hoshi-hiyouga
1344416378
[model] fix moe zero3 (#7826) 2025-04-23 15:30:49 +08:00
Kingsley
1dd67eb042
[data] fix internvl plugin (#7817) 2025-04-23 00:58:22 +08:00
hoshi-hiyouga
2b7d564e3b
[assets] update model readme (#7804) 2025-04-22 16:43:56 +08:00
Kingsley
d43013f14a
[model] add arch check for InternVL (#7803) 2025-04-22 16:38:05 +08:00
Kingsley
c91165a5a6
[misc] update internvl constants (#7801) 2025-04-22 15:53:08 +08:00
hoshi-hiyouga
7f3c31f6f4
[trainer] support early stop (#7797) 2025-04-22 01:59:33 +08:00
hoshi-hiyouga
92101f34a1
[data] improve mmplugin (#7795) 2025-04-22 01:25:33 +08:00
hoshi-hiyouga
a62cba3d05
[example] add bash usage (#7794) 2025-04-22 00:25:51 +08:00
Juanxi Tian
d128382d3c
[trainer] Add Muon Optimizer (#7749)
Co-authored-by: hoshi-hiyouga <hiyouga@buaa.edu.cn>
2025-04-21 23:38:37 +08:00
hoshi-hiyouga
278df4308d
[parser] support omegaconf (#7793) 2025-04-21 23:30:30 +08:00
Changrui Chen
81768df04c
[data] Fix wrong position ids with packed attention masks (#7754)
Co-authored-by: hoshi-hiyouga <hiyouga@buaa.edu.cn>
2025-04-21 23:19:36 +08:00
flashJd
1302ca39f6
[misc] fix new tokens adding (#7253)
Co-authored-by: hoshi-hiyouga <hiyouga@buaa.edu.cn>
2025-04-21 23:19:02 +08:00
ddddng
b8cddbc7d7
[model] fix gemma3 export (#7786)
Co-authored-by: hoshi-hiyouga <hiyouga@buaa.edu.cn>
2025-04-21 23:07:11 +08:00
Sachin Beldona
ec7257e70f
[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
a4455e3021
[assets] update wechat (#7792) 2025-04-21 21:29:42 +08:00
hoshi-hiyouga
610f164c69
[trainer] fix pt loss (#7748)
* fix pt loss

* robust

* fix

* test
2025-04-17 03:15:35 +08:00
hoshi-hiyouga
0a0cfeb782
[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
4831552856
[infer] set env for vllm ascend (#7745) 2025-04-17 01:08:55 +08:00
Kingsley
125513fa5c
[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
ENg-122
8543400584
[misc] improve entrypoint (#7345)
* 纯粹优化下入口代码,因为看到if else太多了

* Update cli.py

---------

Co-authored-by: hoshi-hiyouga <hiyouga@buaa.edu.cn>
2025-04-16 21:48:23 +08:00
leo-pony
e1fdd6e2f8
[infer] support vllm-ascend (#7739) 2025-04-16 20:06:47 +08:00
codingma
d07983dceb
[assets] wechat (#7740) 2025-04-16 18:02:01 +08:00
hoshi-hiyouga
9b94211045
[api] fix chat messages (#7732) 2025-04-15 16:39:08 +08:00
hoshi-hiyouga
0fe5631f9b
[deps] upgrade vllm (#7728) 2025-04-15 14:57:40 +08:00
Joe Schoonover
b5d667cebf
[docker] patch docker-rocm (#7725)
* Update Dockerfile

* Fix typo

* Fix syntax for /bin/sh conditional

* Add build args to docker-compose

* Change shell to /bin/bash

This is required for "==" syntax in conditional string comparison
2025-04-15 13:36:39 +08:00
hoshi-hiyouga
ac8c6fdd3a
[assets] update model readme (#7724) 2025-04-15 00:41:09 +08:00
Kingsley
df8752e8ee
[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