Unverified Commit 42fc4410 authored by Ying Sheng's avatar Ying Sheng Committed by GitHub
Browse files

[minor] Add server_args check for Llama4 with hybrid (#7988)

parent 5f6756b0
......@@ -1730,6 +1730,10 @@ class ServerArgs:
else:
self.lora_paths[lora_path] = lora_path
model_arch = get_model_arch(self)
if "Llama4" in model_arch and self.hybrid_kvcache_ratio is not None:
assert self.attention_backend == "fa3"
def prepare_server_args(argv: List[str]) -> ServerArgs:
"""
......
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