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

This commit is contained in:
Ying Sheng
2025-07-12 20:13:40 -07:00
committed by GitHub
parent 5f6756b038
commit 42fc44100a

View File

@@ -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:
"""