A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 13, 23:15 UTC
feat(benchmarks): support HF model names in multi-turn benchmark (#27850)
ai-jz committed
Nov 1, 2025 at 01:04 UTC
2c0c7c39bdf78ff4cf99a93f67066435e1712cd8
1 file changed
-2
benchmarks/multi_turn/benchmark_serving_multi_turn.py
-2
@@ -1429,8 +1429,6 @@ async def main() -> None:
1429
random.seed(args.seed)
1430
np.random.seed(args.seed)
1431
1432
- if not os.path.exists(args.model):
1433
- raise OSError(f"Path does not exist: {args.model}")
1432
logger.info("Loading tokenizer")
1433
tokenizer = AutoTokenizer.from_pretrained(args.model)
1434