Unverified Commit ab5666eb authored by Matthew Bonanni's avatar Matthew Bonanni Committed by GitHub
Browse files

[UX] Bump version in CG memory profiling log message (#40465)


Signed-off-by: default avatarMatthew Bonanni <mbonanni@redhat.com>
parent f819265a
......@@ -456,7 +456,7 @@ class Worker(WorkerBase):
logger.info(
"CUDA graph memory profiling is enabled "
"(VLLM_MEMORY_PROFILER_ESTIMATE_CUDAGRAPHS=1). "
"This will become the default in v0.19. "
"This will become the default in v0.21. "
"The current --gpu-memory-utilization=%.4f is equivalent "
"to --gpu-memory-utilization=%.4f without CUDA graph "
"memory profiling. To maintain the same effective KV "
......@@ -472,7 +472,7 @@ class Worker(WorkerBase):
1.0,
)
logger.info(
"In v0.19, CUDA graph memory profiling will be enabled "
"In v0.21, CUDA graph memory profiling will be enabled "
"by default (VLLM_MEMORY_PROFILER_ESTIMATE_CUDAGRAPHS=1), "
"which more accurately accounts for CUDA graph memory "
"during KV cache allocation. To try it now, set "
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment