ip

Author: oandreeva-nv

ci/L0_backend_vllm/accuracy_test/test.sh

@@ -48,10 +48,10 @@ RET=0
 set +e
 # Need to generate baseline first, since running 2 vLLM engines causes
 # memory issues: https://github.com/vllm-project/vllm/issues/2248
-VLLM_USE_V1=0 python3 $CLIENT_PY --generate-baseline >> $VLLM_ENGINE_LOG 2>&1 & BASELINE_PID=$!
+VLLM_USE_V1=0 VLLM_WORKER_MULTIPROC_METHOD=spawn python3 $CLIENT_PY --generate-baseline >> $VLLM_ENGINE_LOG 2>&1 & BASELINE_PID=$!
 wait $BASELINE_PID
 
-VLLM_USE_V1=0 python3 $CLIENT_PY --generate-guided-baseline > $VLLM_ENGINE_LOG 2>&1 & BASELINE_PID=$!
+VLLM_USE_V1=0 VLLM_WORKER_MULTIPROC_METHOD=spawn python3 $CLIENT_PY --generate-guided-baseline > $VLLM_ENGINE_LOG 2>&1 & BASELINE_PID=$!
 wait $BASELINE_PID
 
 set -e