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 10, 20:15 UTC
[Bugfix]Fix pooling model always disabled due to incorrect PP rank check (#31505)
Signed-off-by: vintipandey <vinti.pandey@gmail.com>
vintipandey committed
Dec 30, 2025 at 11:27 UTC
04147dcfa70fb7228ce9e2f88fa7dd41631d17f0
1 file changed
+1
-1
vllm/v1/worker/gpu_model_runner.py
+1
-1
@@ -326,7 +326,7 @@ class GPUModelRunner(
326
# https://github.com/vllm-project/vllm/issues/18019
327
self.broadcast_pp_output = (
328
self.parallel_config.distributed_executor_backend == "external_launcher"
329
- and len(get_pp_group().ranks) > 0
329
+ and len(get_pp_group().ranks) > 1
330
)
331
332
# Model-related.