Revise the default logic for the model cache RAM limit #7566
Conversation
This file contains 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
github-actions
bot
added
python
RyanJDick
force-pushed
the
ryan/lower-virtual-memory-3
branch
from
9235ede to
ce57c4e
Compare
14 tasks
psychedelicious
approved these changes
Jan 17, 2025
|
Code review looks good. Let me know if I can help with testing. |
RyanJDick
requested review from
lstein,
blessedcoolant,
brandonrising and
hipsterusername
as code owners
hipsterusername
approved these changes
Jan 17, 2025
RyanJDick
added a commit
that referenced
this pull request
Jan 17, 2025
## Summary This PR adds a `keep_ram_copy_of_weights` config option the default (and legacy) behavior is `true`. The tradeoffs for this setting are as follows: - `keep_ram_copy_of_weights: true`: Faster model switching and LoRA patching. - `keep_ram_copy_of_weights: false`: Lower average RAM load (may not help significantly with peak RAM). ## Related Issues / Discussions - Helps with #7563 - The Low-VRAM docs are updated to include this feature in #7566 ## QA Instructions - Test with `enable_partial_load: false` and `keep_ram_copy_of_weights: false`. - [x] RAM usage when model is loaded is reduced. - [x] Model loading / unloading works as expected. - [x] LoRA patching still works. - Test with `enable_partial_load: false` and `keep_ram_copy_of_weights: true`. - [x] Behavior should be unchanged. - Test with `enable_partial_load: true` and `keep_ram_copy_of_weights: false`. - [x] RAM usage when model is loaded is reduced. - [x] Model loading / unloading works as expected. - [x] LoRA patching still works. - Test with `enable_partial_load: true` and `keep_ram_copy_of_weights: true`. - [x] Behavior should be unchanged. - [x] Smoke test CPU-only and MPS with default configs. ## Merge Plan - [x] Merge #7564 first and change target branch. ## Checklist - [x] _The PR has a short but descriptive title, suitable for a changelog_ - [x] _Tests added / updated (if applicable)_ - [ ] _Documentation added / updated (if applicable)_ - [ ] _Updated `What's New` copy (if doing a release after this PR)_
|
@RyanJDick I wonder if we should figure out special handling for MPS devices and their unified memory architecture. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Summary
This PR revises the logic for calculating the model cache RAM limit. See the code for thorough documentation of the change.
The updated logic is more conservative in the amount of RAM that it will use. This will likely be a better default for more users. Of course, users can still choose to set a more aggressive limit by overriding the logic with
max_cache_ram_gb.Related Issues / Discussions
QA Instructions
Exercise all heuristics:
Merge Plan
keep_ram_copy_of_weightsconfig option #7565 first and update the target branchChecklist
What's Newcopy (if doing a release after this PR)