diff --git a/vllm_omni/worker/cpu/cpu_ar_model_runner.py b/vllm_omni/worker/cpu/cpu_ar_model_runner.py index f4ce7f5dd..70a8fbbb4 100644 --- a/vllm_omni/worker/cpu/cpu_ar_model_runner.py +++ b/vllm_omni/worker/cpu/cpu_ar_model_runner.py @@ -1,4 +1,4 @@ -"""AR GPU Model Runner for vLLM-omni. +"""AR CPU Model Runner for vLLM-omni. Exposes per-request hidden representations via ModelRunnerOutput.pooler_output and also outputs sampled tokens.