[V0 Deprecation] Remove max_seq_len_to_capture (#25543)
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
This commit is contained in:
@@ -31,7 +31,6 @@ def use_v1_only(monkeypatch: pytest.MonkeyPatch):
|
||||
def setup_vllm(num_loras: int, tp: int) -> vllm.LLM:
|
||||
return vllm.LLM(model="Qwen/Qwen2.5-3B-Instruct",
|
||||
max_model_len=256,
|
||||
max_seq_len_to_capture=256,
|
||||
max_num_seqs=8,
|
||||
tensor_parallel_size=tp,
|
||||
enable_lora=True,
|
||||
|
||||
Reference in New Issue
Block a user