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 7, 11:30 UTC
[BUG] Fix rlhf_async example (#35788)
Signed-off-by: ahao-anyscale <ahao@anyscale.com>
Aaron Hao committed
Mar 2, 2026 at 12:36 UTC
cad21918e3f1a13353d6fcc1f1f431d6d3baf964
1 file changed
+1
-1
examples/offline_inference/new_weight_syncing/rlhf_async_new_apis.py
+1
-1
@@ -104,7 +104,7 @@ class MyLLM(vllm.AsyncLLMEngine):
104
while not self._request_pause_flag:
105
await asyncio.sleep(0)
106
await super().pause_generation(mode="keep")
107
- await asyncio.sleep(0.2)
107
+ await asyncio.sleep(5)
108
self._generation_paused = True
109
110