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 14, 05:30 UTC
[bugfix]fix empty prompts for async-engine mode in benchmark throughput (#27494)
Signed-off-by: Lucia Fang <fanglu@fb.com>
Lucia Fang committed
Oct 26, 2025 at 01:16 UTC
315b860abe93634aa0a7db4aa3468f5441786972
1 file changed
+1
vllm/benchmarks/throughput.py
+1
@@ -221,6 +221,7 @@ async def run_vllm_async(
221
detokenize=not disable_detokenize,
222
)
223
)
224
+ prompts.append(prompt)
225
lora_requests.append(request.lora_request)
226
227
generators = []