fix: simplify determine_available_memory#317
Merged
rebel-jaehwang merged 1 commit intodev-0.12from Jan 30, 2026
Merged
Conversation
rebel-jaehwang
force-pushed
the
avail-mem
branch
from
ab28a18 to
a58f24b
Compare
* We don't need to compute the number of blocks since vllm allocator already does it, properly considering diffferent layer types. * Remove VLLM_RBLN_NPU_NUM_BLOCKS. User should use the standard gpu_memory_utilization config instead. * Don't take min with max of memory used for active request, since prefix cache can utilize the extra memory.
rebel-jaehwang
force-pushed
the
avail-mem
branch
from
a58f24b to
5d8bd7a
Compare
rebel-jaehwang
added a commit
that referenced
this pull request
Jan 30, 2026
* We don't need to compute the number of blocks since vllm allocator already does it, properly considering diffferent layer types. * Remove VLLM_RBLN_NPU_NUM_BLOCKS. User should use the standard gpu_memory_utilization config instead. * Don't take min with max of memory used for active request, since prefix cache can utilize the extra memory.
rebel-jaehwang
added a commit
that referenced
this pull request
Jan 30, 2026
* We don't need to compute the number of blocks since vllm allocator already does it, properly considering diffferent layer types. * Remove VLLM_RBLN_NPU_NUM_BLOCKS. User should use the standard gpu_memory_utilization config instead. * Don't take min with max of memory used for active request, since prefix cache can utilize the extra memory.
rebel-jaehwang
added a commit
that referenced
this pull request
Jan 30, 2026
* We don't need to compute the number of blocks since vllm allocator already does it, properly considering diffferent layer types. * Remove VLLM_RBLN_NPU_NUM_BLOCKS. User should use the standard gpu_memory_utilization config instead. * Don't take min with max of memory used for active request, since prefix cache can utilize the extra memory.
rebel-jiwoopark
pushed a commit
that referenced
this pull request
Feb 4, 2026
* We don't need to compute the number of blocks since vllm allocator already does it, properly considering diffferent layer types. * Remove VLLM_RBLN_NPU_NUM_BLOCKS. User should use the standard gpu_memory_utilization config instead. * Don't take min with max of memory used for active request, since prefix cache can utilize the extra memory.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
already does it, properly considering diffferent layer types.
gpu_memory_utilization config instead.
prefix cache can utilize the extra memory.