xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-15 13:55:38 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix#30092 Kimi-Linear model loading failure with missing indexer_rotary_emb (#30093)

Browse Source 
Signed-off-by: baonudesifeizhai <baonudesifeizhai@gmail.com>
This commit is contained in:
Dongjie Zou 2025-12-05 23:55:09 -05:00 committed by GitHub
parent c4d62618ca
commit e3fbb6f152
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/model_executor/layers/mla.py
View File

@ -24,9 +24,9 @@ class MLAModules:
q_b_proj: torch.nn.Module | None q_b_proj: torch.nn.Module | None
q_proj: torch.nn.Module | None q_proj: torch.nn.Module | None
indexer: torch.nn.Module | None indexer: torch.nn.Module | None
indexer_rotary_emb: torch.nn.Module | None
is_sparse: bool is_sparse: bool
topk_indices_buffer: torch.Tensor | None topk_indices_buffer: torch.Tensor | None
indexer_rotary_emb: torch.nn.Module | None = None
@CustomOp.register("multi_head_latent_attention") @CustomOp.register("multi_head_latent_attention")