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 15, 11:00 UTC
Fix broken example: examples/offline_inference/profiling at scheduler_config (#18117)
Ecthlion_zyy committed
May 14, 2025 at 14:19 UTC
33011318c20824799b817347783d911e5b2076ea
1 file changed
+1
-1
examples/offline_inference/profiling.py
+1
-1
@@ -193,7 +193,7 @@ def run_profile(context: ProfileContext, csv_output: Optional[str],
193
batch_size = context.batch_size
194
prompt_len = context.prompt_len
195
196
- scheduler_config = llm.llm_engine.scheduler_config
196
+ scheduler_config = llm.llm_engine.vllm_config.scheduler_config
197
max_model_len = llm.llm_engine.model_config.max_model_len
198
max_num_batched_tokens = scheduler_config.max_num_batched_tokens
199
max_num_seqs = scheduler_config.max_num_seqs