Shanshan Shen 89ca63a2c2 [Bugfix] Disable torch.compile() (#370) ...

### What this PR does / why we need it?
To resolve this
[patch](https://github.com/vllm-project/vllm-ascend/pull/236/files#diff-43b96b39b5a52fe209d86449ad703a7ff5e1349ebaf1aa12ece8d82163ee5b61R24-R49)
, we need to set `torch.compile()` backend to `eager` to disable
compile, using default pytorch way.


---------

Signed-off-by: shen-shanshan <467638484@qq.com>

2025-03-21 15:55:51 +08:00

..

attention

[Bugfix] Disable torch.compile() (#370)

2025-03-21 15:55:51 +08:00

models

bugfix for qwen2_vl (#301)

2025-03-12 08:39:50 +08:00

ops

[Feature] Implement EP-compatible fused_moe (#121)

2025-03-11 21:08:02 +08:00

quantization

[BugFix] Fix bugs when using ascend quantization (#275)

2025-03-12 11:33:21 +08:00

worker

[Bugfix] Disable torch.compile() (#370)

2025-03-21 15:55:51 +08:00

__init__.py

register qwen2_vl to rewrite qwen2_vl forwad (#241)

2025-03-07 15:41:47 +08:00

communicator.py

[Dist] Set device as rank (#202)

2025-03-03 09:23:13 +08:00

platform.py

[Bugfix] Disable torch.compile() (#370)

2025-03-21 15:55:51 +08:00

utils.py

support multistep decode (#299)

2025-03-11 19:20:06 +08:00