Enable Chroma finetune within 32gb RAM & 24gb VRAM #488
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
stefanasandei
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
Hello, I was trying to finetune Chroma1-HD on my setup, RTX 3090 and 32gb RAM. However, I was getting stuck during model loading, because the RAM was getting filled, resulting in my computer freezing every time.
In the
ChromaModelclass, the safetensors file is loaded first into the CPU, to be quantized. The Chroma checkpoint is too big to be loaded in full precision within 32gb RAM, so I changed the global dtype to bebfloat16during the initial loading, and then the normal dtypes afterwards.With this configuration, I can easily finetune Chroma with my setup.
I also fixed a bug related to
FakeTextEncoderduring loading (related issue: #405).