423A35C7

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-08-14 03:55:44 +08:00

f28afaf635 [v1] add FSDPTurbo EP/EFSDP plugin for MoE training (#10676)

bc4b42cefc [train] Harden KTransformers MoE LoRA SFT integration (#10738)

199b8873d7 [train] add fa3 supported (#10742)

Compare 3 commits »

423A35C7 synced commits to main at 423A35C7/pytorch3d from mirror 2026-08-14 03:45:43 +08:00

3143b3baf8 Read implicitron's own annotations via inspect.get_annotations

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-08-10 18:15:44 +08:00

0bbe481e6e [docker] upgrade NPU images to CANN 9.1 and PyTorch 2.10 (#10729)

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-08-09 17:45:45 +08:00

63a89710c7 [data] pad position_ids on non-FA2 packing path (fixes rotary crash for Gemma-3/4) (#10737)

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-08-06 16:16:10 +08:00

887b850813 [assets] fix broken links in README (#10728)

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-08-04 23:26:10 +08:00

84576b1408 [v1] refactor NPU kernel matching by model type (#10643)

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-08-03 22:56:11 +08:00

713b5a3f95 [model] add MOSS-VL support (#10708)

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-07-31 21:26:09 +08:00

62ae362455 [v1] Support multimodal data training (#10656)

3984675dd5 fix(ci): align workflow Python version with requires-python (#10707)

Compare 2 commits »

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-07-30 20:56:10 +08:00

1b47415a2f [train] Fix hyper parallel tail accumulation loss scaling (#10705)

423A35C7 synced commits to main at 423A35C7/pytorch3d from mirror 2026-07-29 12:06:09 +08:00

9381c40163 Suppress type errors for Pyre upgrade

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-07-27 19:26:09 +08:00

9ce6b663e9 [train] support megatron-bridge for PT/SFT training (#10645)

423A35C7 synced commits to main at 423A35C7/pytorch3d from mirror 2026-07-25 10:06:09 +08:00

32a33e2442 Enable Pyrefly in fbcode/vision/fair/pytorch3d

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-07-24 17:56:09 +08:00

2ebe7be611 [ci] pin ruff version and fix lint errors (#10681)

3f77101580 [v1] refactor registry plugin structure and params (#10641)

19e9fe3ced [docker] improve NPU image build and distribution (#10664)

Compare 3 commits »

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-07-24 01:36:09 +08:00

d0eaa10b0c [docs] update readme (#10678)

a17afe5e1b [docs] update trend badge and promote PenguinHarness in readme (#10677)

Compare 2 commits »

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-07-18 06:46:09 +08:00

ef2d8f9da6 [v1] fix grad norm and lr log (#10640)

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-07-17 22:36:09 +08:00

5f653cb96a [v1] add muon optimizer (#10618)

423A35C7 synced commits to main at 423A35C7/pytorch3d from mirror 2026-07-16 21:56:09 +08:00

4daa00b41c Fix CQS signal facebook-unused-include-check in fbcode/vision/fair

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-07-15 21:36:09 +08:00

d1049d650a [docs] Add AMD GPU Cloud link (#10649)

423A35C7 synced commits to main at 423A35C7/pytorch3d from mirror 2026-07-15 05:06:10 +08:00

3b5ab51de5 Fix backprop through cot_laplacian

423A35C7 synced commits to main at 423A35C7/LLaMA-Factory from mirror 2026-07-13 20:36:09 +08:00

8489928769 [fix]update license check, update transformers (#10632)