diff --git a/.buildkite/scripts/scheduled_integration_test/deepseek_v2_lite_prefetch_offload.sh b/.buildkite/scripts/scheduled_integration_test/deepseek_v2_lite_prefetch_offload.sh index de48eb282a..0eadfa1f80 100755 --- a/.buildkite/scripts/scheduled_integration_test/deepseek_v2_lite_prefetch_offload.sh +++ b/.buildkite/scripts/scheduled_integration_test/deepseek_v2_lite_prefetch_offload.sh @@ -51,6 +51,7 @@ vllm serve "$MODEL" \ --offload-num-in-group 2 \ --offload-prefetch-step 1 \ --offload-params w13_weight w2_weight \ + --generation-config vllm \ --port "$PORT" \ ${EXTRA_ARGS+"${EXTRA_ARGS[@]}"} & SERVER_PID=$!