"vscode:/vscode.git/clone" did not exist on "ee5667c1ff7bbc0f4edfd661b033b61fc6dbfbec"
Unverified Commit 8e5a6d34 authored by pengcuo's avatar pengcuo Committed by GitHub
Browse files

[Fix] Fix a bug for flashmla to run R1 model (#5875)


Co-authored-by: default avatarpengcuo <dgpengcuo@gmail.com>
parent 8465f035
...@@ -241,6 +241,9 @@ class FlashMLABackend(FlashInferMLAAttnBackend): ...@@ -241,6 +241,9 @@ class FlashMLABackend(FlashInferMLAAttnBackend):
seq_lens_cpu, seq_lens_cpu,
) )
def get_cuda_graph_seq_len_fill_value(self):
return 1024
def forward_decode( def forward_decode(
self, self,
q: torch.Tensor, q: torch.Tensor,
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment