add disable cache in LazyTensorFactory #30
Merged
+11
−1
Conversation
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
Signed-off-by: yuanyuxing.yyx <[email protected]>
ABNER-1
force-pushed
the
add_disable_cache_in_LazyTensorFactory
branch
from
00456d2 to
65ff53f
Compare
takeshi-yoshimura
merged commit 7dfe757
into
foundation-model-stack:main
13 checks passed
Collaborator
|
looks perfect. thanks for contributing! |
This was referenced Dec 2, 2025
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Outdated suggestions cannot be applied.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.
The first in #29
Disables the batch tensor cache by default since it is unnecessary for most scenarios and consumes extra GPU memory.
Example Configuration:
Backend/Framework: SGLang with TP8
Model: deepseek-r1 (each file: 4GiB)
Extra Memory Comparison:
With disable_cache=True: ~5.7 GiB = 4 GiB (file) + 1.7 GiB (largest tensor)
With disable_cache=False: 32 GiB = 4 GiB * 8 (files)
Conclusion: Disabling the cache reduces peak GPU memory usage by approximately 82%.