Remove max_num_seqs in latency benchmark script (#1855)
This commit is contained in:
parent
e74b1736a1
commit
51d3cb951d
@ -19,7 +19,6 @@ def main(args: argparse.Namespace):
|
||||
tokenizer=args.tokenizer,
|
||||
quantization=args.quantization,
|
||||
tensor_parallel_size=args.tensor_parallel_size,
|
||||
max_num_seqs=args.batch_size,
|
||||
trust_remote_code=args.trust_remote_code,
|
||||
dtype=args.dtype,
|
||||
)
|
||||
|
Loading…
x
Reference in New Issue
Block a user