423A35C7

423A35C7 synced and deleted reference refs/tags/hiyouga/acc at 423A35C7/LLaMA-Factory from mirror 2025-12-13 02:16:57 +08:00

423A35C7 synced commits to hiyouga/acc at 423A35C7/LLaMA-Factory from mirror 2025-12-12 18:08:19 +08:00

3b06033357 add accelerator

423A35C7 synced new reference hiyouga/acc to 423A35C7/LLaMA-Factory from mirror 2025-12-12 10:04:22 +08:00

423A35C7 synced commits to hiyouga/acc at 423A35C7/LLaMA-Factory from mirror 2025-12-12 10:04:22 +08:00

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-12-10 17:14:17 +08:00

4fd94141a4 [model] Add Ministral3 (#9582)

22d6ac29d5 [model] Rename GLMV template (#9595)

Compare 2 commits »

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-12-10 00:54:21 +08:00

cff4483392 [config] Fix RoPE scaling patch for resuming from a scaled model (#9588)

5d56817e2b [misc] lint (#9593)

Compare 2 commits »

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-12-09 16:44:20 +08:00

1bbb461f76 [assets] update readme (#9587)

c1f5f8fff6 [model] support GLM4.6v (#9586)

Compare 2 commits »

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-12-08 08:04:17 +08:00

5744f1ea94 [v1] add models & accelerator (#9579)

739954910a [deps] Update for Transformers v5 (#9569)

Compare 2 commits »

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-12-06 23:24:20 +08:00

109162dc56 [fix] fix the issue when using fsdp2 with gradient checkpointing. (#9541)

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-12-06 07:04:18 +08:00

165f3f073a [examples] add fsdp config for mutiple nodes (#9575)

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-12-02 05:04:19 +08:00

efb13b7483 [V1] Refactor ascend MoE kernel patch logic & Support Qwen3-MoE (#9557)

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-11-30 12:14:17 +08:00

e43a972b25 [test] add npu test yaml and add ascend a3 docker file (#9547)

22be45c78c [misc] fix omni thinker load (#9552)

Compare 2 commits »

423A35C7 synced new reference refs/tags/v0.7.9 to 423A35C7/pytorch3d from mirror 2025-11-29 03:54:18 +08:00

423A35C7 synced commits to refs/tags/v0.7.9 at 423A35C7/pytorch3d from mirror 2025-11-29 03:54:18 +08:00

423A35C7 synced commits to main at 423A35C7/pytorch3d from mirror 2025-11-28 03:24:19 +08:00

33824be3cb version 0.7.9

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-11-27 18:54:19 +08:00

d1f585f80a [test] update test cmd (#9544)

955396e8a5 [example] correct the parameter errors in the examples file. (#9543)

Compare 2 commits »

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-11-27 02:34:19 +08:00

231756a5bf [chat] fix the error when the vLLM version is greater than 0.10.0 (#9539)

2c4fb3c97e [v1] Support fused moe kernel for qwen3vlmoe model. (#9532)

2b6f16f261 [model] temporarily support npu fused options on v0, powered by v1 kernels (#9520)

f17efde693 [v1] support automatic discovery of registered kernels. (#9509)

Compare 4 commits »

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-11-24 17:24:18 +08:00

591fc9ed02 [model] support ERNIE-4.5-VL Models (#9521)

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-11-19 23:04:22 +08:00

3140c242f0 [assets] add README with KT+llamafactory (#9514)

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2025-11-19 06:46:35 +08:00

887c562d60 [example] Add KTransformers Qwen3MoE example (#9511)