Skip to content

[CLI] Report KV cache memory usage in mlc_llm compile #3221

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

[CLI] Report KV cache memory usage in mlc_llm compile #3221

wants to merge 2 commits into from
+38 −1

Conversation

CharlieFRuan
Copy link
Contributor

@CharlieFRuan CharlieFRuan commented May 4, 2025 •
edited
Loading

This PR prints out the memory usage of KV cache: MB for one token's KV cache, and the total MB for model weights + intermediate buffers + a 4K-long KV cache.

If somehow the required fields are not present in config and metadata (e.g. for an old model), we do nothing.

Sample output in CLI:

[2025-05-03 21:44:24] INFO model_metadata.py:94: Total memory usage without KV cache: 2254.16 MB (Parameters: 923.16 MB. Temporary buffer: 1331.00 MB)
[2025-05-03 21:44:24] INFO model_metadata.py:128: KV cache size: 0.11 MB per token in the context window
[2025-05-03 21:44:24] INFO model_metadata.py:133: Total memory usage with a 4K KV cache: 2702.16 MB

@CharlieFRuan CharlieFRuan requested a review from MasterJH5574 May 4, 2025 02:00
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@MasterJH5574 MasterJH5574 Awaiting requested review from MasterJH5574

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@CharlieFRuan