2813 Commits

Author SHA1 Message Date
2c6aded5d4 [v1] kernel plugin (#9274)
Co-authored-by: frozenleaves <frozen@Mac.local>
2025-10-18 18:02:14 +08:00
d9d67ba62d [misc] fix import error (#9299) 2025-10-17 17:46:27 +08:00
a442fa90ad [misc] fix import error (#9296) 2025-10-17 10:54:30 +08:00
8c341cbaae [model] support hunyuan-mt model (#9284)
Co-authored-by: wyfdgg <liwenkun0812@163.com>
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>
2025-10-17 10:33:09 +08:00
47a7dc1698 [deps] upgrade vllm (#9293) 2025-10-16 23:20:26 +08:00
1037f63311 [model] add qwen3vl 4b + 8b (#9275) 2025-10-15 15:00:36 +08:00
c867e28093 [model] adds semantic initialization support for special tokens (#9267)
Co-authored-by: ximingxing <ximingxing@tencent.com>
2025-10-14 17:00:48 +08:00
3dbca4b533 [data] add new reason tool calls demo data (#9249)
Co-authored-by: unknown <Peter Zeng>
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>
2025-10-13 17:16:47 +08:00
9d1acbc191 [ci] fix ci (#9265) 2025-10-13 16:24:40 +08:00
52e46e162e [v1] add data converter (#9263) 2025-10-13 15:54:47 +08:00
48974783da [model]: add ernie4_5_moe support for DeepSpeed Zero3 training (#9262) 2025-10-13 13:13:31 +08:00
575e4099df [misc] add qwen bench script (#9259)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-10-13 11:45:25 +08:00
9687b71d3a [v1] init data plugins (#9248)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-10-09 22:36:48 +08:00
1c35db60d6 [v1] support read dataset (#9243) 2025-10-09 17:16:33 +08:00
10146029ba [v1] add v1 launcher (#9236)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-10-07 22:34:48 +08:00
95b7188090 Merge commit from fork
* fix lfi and ssrf

* move utils to common

---------

Co-authored-by: d3do <chamlinx@outlook.com>
Co-authored-by: hiyouga <hiyouga@buaa.edu.cn>
2025-10-07 20:55:29 +08:00
d5bb4e6394 [assets] update readme (#9232) 2025-10-05 16:42:19 +08:00
3fe6f0febd [ci] update docker workflow (#9231) 2025-10-05 02:50:55 +08:00
40d3691e9e [misc] fix moe models (#9230) 2025-10-05 02:41:02 +08:00
af8437095a [ci] Change macOS version (#9229) 2025-10-05 02:18:30 +08:00
2e2f92701f [model] add qwen3-vl-30b (#9227) 2025-10-04 14:12:37 +08:00
7d60b840ef [v1] support switch v1 backend (#9226) 2025-10-02 15:59:19 +08:00
1d96c62df2 [v1] add v1 folders (#9225) 2025-10-02 15:25:57 +08:00
a0d44c650a [misc] add data files (#9224) 2025-10-02 14:02:07 +08:00
bcc2c1fd8f [misc] move wechat out (#9223) 2025-10-02 02:06:09 +08:00
7dd910f067 [misc] lint (#9221) 2025-10-01 22:58:58 +08:00
d10d65e4ce [docker] update Dockerfile to set no_proxy and fix pydantic version (#8651) 2025-10-01 14:33:47 +08:00
1c44b60e3e [feat] fp8 training (#8960)
Co-authored-by: Benjamin Feuer <penfever@gmail.com>
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>
2025-10-01 14:32:53 +08:00
e2b1594d31 [data] fix reasoning template (#9219) 2025-09-30 18:11:45 +08:00
09dedf144f [npu] Redirect SDPA to torch_npu.npu_fusion_attention (opt-in, ZeRO-3 safe, no impact off NPU) (#8972) 2025-09-30 18:11:31 +08:00
a04d777d7f [cli] support lazy import (#9217)
Co-authored-by: frozenleaves <frozen@Mac.local>
2025-09-30 18:02:26 +08:00
6ffebe5ff7 [data] fix qwen omni plugin (#9204)
Co-authored-by: kingsley <kingsleydodonow@gmail.com>
2025-09-28 01:02:29 +08:00
0761a4448f [model] add qwen3-vl/qwen3-omni (#9196)
Co-authored-by: kingsley <kingsleydodonow@gmail.com>
2025-09-27 01:21:47 +08:00
abc3b1e1c4 [docs] update ling-v2 to the readme (#9188) 2025-09-24 15:23:21 +08:00
344c760cc1 [model] supported ERNIE4.5 Text Models (#9165)
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>
2025-09-22 11:48:26 +08:00
80fe3a172d [model] add dots ocr (#9176) 2025-09-21 23:34:19 +08:00
800934b507 [assets] update readme (#9143) 2025-09-16 17:04:19 +08:00
e2ba32598d [assets] update readme (#9137) 2025-09-15 23:45:57 +08:00
812720909e [model] add qwen3 next (#9130) 2025-09-14 03:16:25 +08:00
260b5625c3 [assets] update wechat (#9129) 2025-09-14 03:05:08 +08:00
52488ac974 [deps] upgrade transformers to 4.56.1 (#9128) 2025-09-14 02:26:39 +08:00
610a3f1094 [data] Fix qwen_2vl with valuehead (#9078) 2025-09-14 02:22:20 +08:00
a22dab97fd [data] bailing template v2 & openai data converter (#9112) 2025-09-14 02:20:34 +08:00
db223e3975 [misc] update readme (#9071) 2025-09-03 17:22:54 +08:00
7e710c6d3e [misc] update InternVL constants (#9046) 2025-08-29 13:30:28 +08:00
185f0556d4 [model] support Internvl3_5 (#9028) 2025-08-28 17:12:00 +08:00
1c675522fd [data] Fix glm4v mmplugin when not expand (#9029) 2025-08-27 11:45:24 +08:00
6c777375b7 [model] Support MiniCPM-V 4.5 (#9022) 2025-08-26 19:27:28 +08:00
9c433f6b41 [model] fix kimivl (#9018) 2025-08-25 16:32:23 +08:00
ec41ef08aa [misc] fix constants (#9008) 2025-08-23 23:04:30 +08:00