Damien Benveniste commited on
Commit
9499c26
·
1 Parent(s): 9ed86a1
Files changed (1) hide show
  1. entrypoint.sh +1 -1
entrypoint.sh CHANGED
@@ -34,7 +34,7 @@ CMD="vllm serve $MODEL \
34
  --max-num-batched-tokens $MAX_NUM_BATCHED_TOKENS \
35
  --max-num-seqs $MAX_NUM_SEQS \
36
  --gpu-memory-utilization $GPU_MEMORY_UTILIZATION \
37
- --max-model-len $MAX_MODEL_LEN
38
 
39
  # Add enforce-eager only if it's set to true
40
  if [ "$ENFORCE_EAGER" = "true" ]; then
 
34
  --max-num-batched-tokens $MAX_NUM_BATCHED_TOKENS \
35
  --max-num-seqs $MAX_NUM_SEQS \
36
  --gpu-memory-utilization $GPU_MEMORY_UTILIZATION \
37
+ --max-model-len $MAX_MODEL_LEN"
38
 
39
  # Add enforce-eager only if it's set to true
40
  if [ "$ENFORCE_EAGER" = "true" ]; then