From 6462feef65f43ed47b24c76c3f4331ffadcf9644 Mon Sep 17 00:00:00 2001 From: Wentao Ye <44945378+yewentao256@users.noreply.github.com> Date: Tue, 23 Sep 2025 12:44:37 -0400 Subject: [PATCH] [Log] Optimize kv cache memory log from Bytes to GiB (#25204) Signed-off-by: yewentao256 --- vllm/v1/worker/gpu_worker.py | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/vllm/v1/worker/gpu_worker.py b/vllm/v1/worker/gpu_worker.py index ffea9bb35513a..f59dacf13d85a 100644 --- a/vllm/v1/worker/gpu_worker.py +++ b/vllm/v1/worker/gpu_worker.py @@ -383,11 +383,13 @@ class Worker(WorkerBase): f"for non-torch memory, and {GiB(cuda_graph_memory_bytes)} " f"GiB for CUDAGraph memory. Replace gpu_memory_utilization " f"config with `--kv-cache-memory=" - f"{kv_cache_memory_bytes_to_requested_limit}` to fit into " - f"requested memory, or `--kv-cache-memory=" - f"{kv_cache_memory_bytes_to_gpu_limit}` to fully " + f"{kv_cache_memory_bytes_to_requested_limit}` " + f"({GiB(kv_cache_memory_bytes_to_requested_limit)} GiB) to fit " + f"into requested memory, or `--kv-cache-memory=" + f"{kv_cache_memory_bytes_to_gpu_limit}` " + f"({GiB(kv_cache_memory_bytes_to_gpu_limit)} GiB) to fully " f"utilize gpu memory. Current kv cache memory in use is " - f"{int(self.available_kv_cache_memory_bytes)} bytes.") + f"{GiB(self.available_kv_cache_memory_bytes)} GiB.") logger.debug(msg)