2965 Commits

Author SHA1 Message Date
321444ef78 dev npu fused options 2025-09-11 09:23:46 +08:00
eafa225c97 dev npu fused options 2025-09-10 10:35:20 +08:00
cc6ccdb626 dev npu fused options 2025-09-03 16:33:17 +08:00
715834cc16 dev npu fused options 2025-09-03 16:22:21 +08:00
1614a15329 dev npu fused options 2025-08-25 16:31:35 +08:00
d3791e8ee1 [model] Support Intern-S1-mini (#8976) 2025-08-20 23:52:51 +08:00
bd3c197c8f [misc] update glm4v ligerkernel (#8978) 2025-08-20 23:39:56 +08:00
de5ce7da86 [assets] update wechat (#8975) 2025-08-20 09:40:28 +08:00
ac0f50a09a [misc] fix gemma template (#8970) 2025-08-20 03:18:18 +08:00
44beb76cdf [assets] update wechat (#8962) 2025-08-19 02:55:09 +08:00
842595698b [feature] adding orthogononal finetuning (OFT) to llama factory (#8623)
Co-authored-by: Zeju <zqiu@g003.internal.cluster.is.localnet>
Co-authored-by: Zeju <zqiu@login2.is.localnet>
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>
2025-08-18 18:22:47 +08:00
0454c10456 [feature] add support for dft loss (#8917) 2025-08-15 23:29:57 +08:00
2b66b4df43 [feature] Support MPO (#8930) 2025-08-15 15:09:59 +08:00
e557e71023 [data] Fix MiniCPMO default system prompt (#8906) 2025-08-13 23:05:20 +08:00
53a9924ea8 [assets] update wechat & readme (#8899) 2025-08-12 22:45:55 +08:00
a0e9a36a6f [model] add qwen3 nothink (#8869) 2025-08-11 23:17:32 +08:00
8efa5c9694 [model] support GLM4.5V (#8876) 2025-08-11 21:45:14 +08:00
b2395b25b0 [hparams] fix data args (#8863) 2025-08-08 15:35:50 +08:00
df364998b1 [file] Resolve file lock issue when deleting safetensors on Windows (#8839) 2025-08-08 14:59:54 +08:00
d8e4482849 [model] add qwen3-4b-i/t (#8851) 2025-08-07 19:39:26 +08:00
2572111616 [data-loader] Allow dataset_dir to accept a dict for in-memory dataset_info (#8845) 2025-08-07 16:26:59 +08:00
bc54ed8efb [data] fix template (#8827) 2025-08-06 06:58:09 +08:00
706b3e5ee7 [model] add gpt oss (#8826) 2025-08-06 05:56:46 +08:00
48615ddb07 [train] fix adjusting logits size after adding special tokens (#8823) 2025-08-05 20:35:07 +08:00
b3445a75f0 [assets] update readme & supporters (#8818) 2025-08-05 11:00:53 +08:00
42afe1ddd5 [model] Support MiniCPM-V 4.0 (#8813) 2025-08-04 12:57:21 +08:00
4ba7de0434 [model] Qwen3-30B-A3B-Thinking-2507 (#8788) 2025-07-30 23:05:12 +08:00
ea8a2d60d0 [deps] upgrade vllm to 0.10.0 (#8787) 2025-07-30 22:26:38 +08:00
ae0ef374a3 [assets] update readme (#8784) 2025-07-30 17:57:17 +08:00
edd112f35c [model] add qwen3 2507 model (#8783) 2025-07-30 17:19:19 +08:00
7218d4aa96 [model] support keye-vl-8b (#8776) 2025-07-29 21:24:08 +08:00
4380b7b35e [model] update glm4.5 (#8770) 2025-07-29 19:57:29 +08:00
3307ff1d4a [model] add qwen3 2507 models (#8750) 2025-07-25 20:21:47 +08:00
2aadc90c2d [model] add glm4moe (#8689) 2025-07-25 19:53:45 +08:00
2353e16e20 [assets] update readme (#8739) 2025-07-24 23:13:24 +08:00
6812f5e1f5 [assets] update wechat (#8731) 2025-07-24 02:35:47 +08:00
2077875622 [model] support granite4 (#8680) 2025-07-21 14:15:36 +08:00
678b7d69d2 [assets] update wechat (#8685) 2025-07-18 19:13:26 +08:00
f00742b078 [assets] update wechat (#8637) 2025-07-14 21:24:28 +08:00
fdb70c04e0 [data] fix gemma3n mmplugin (#8627) 2025-07-14 13:32:57 +08:00
95ed6c45cd [model] add Devstral-Small-2507 (#8614) 2025-07-11 18:59:53 +08:00
cf1087d409 [webui] fix elems (#8587) 2025-07-08 21:16:42 +08:00
766884fa5c [data] support glm4.1v video training (#8571) 2025-07-08 16:29:44 +08:00
6a8d88826e [webui] fix abort finish (#8569) 2025-07-07 23:07:46 +08:00
043103e1c9 [webui] support other hub (#8567) 2025-07-07 22:18:48 +08:00
5817583630 [deps] bump transformers to 4.49.0 (#8564) 2025-07-07 20:31:50 +08:00
62bd2c8047 [assets] update wechat (#8565) 2025-07-07 20:29:20 +08:00
1b549e3199 [docs] add nvidia-container-toolkit to Linux Docker setup instructions (#8557) 2025-07-06 19:37:08 +08:00
c6290db118 Revert "[model] add lora dropout to unsloth" - requested feature already exists (#8554)
Co-authored-by: viyer <vivek_iyer2@apple.com>
2025-07-05 11:25:31 +08:00
d30cbcdfa5 [ci] Add workflow for building NPU image (#8546)
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>
2025-07-04 20:56:59 +08:00