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 16, 10:45 UTC
[Bugfix] Fix priority in multiprocessing engine (#9277)
Sebastian Schoennenbeck committed
Oct 11, 2024 at 17:35 UTC
df3dcdf49dccfa4914d825fa08b74de8ae050e1e
1 file changed
+2
-1
vllm/engine/multiprocessing/engine.py
+2
-1
@@ -285,7 +285,8 @@ class MQLLMEngine:
285
params=request.params,
286
lora_request=request.lora_request,
287
trace_headers=request.trace_headers,
288
- prompt_adapter_request=request.prompt_adapter_request)
288
+ prompt_adapter_request=request.prompt_adapter_request,
289
+ priority=request.priority)
290
291
if self.log_requests:
292
logger.info("Added request %s.", request.request_id)