Skip to content

Conversation

@sayakpaul
Copy link
Member

What does this PR do?

Just a small QoL improvement for pipeline-level quantization.

PipelineQuantizationConfig(
    quant_backend="bitsandbytes_4bit",
    quant_kwargs={"load_in_4bit": True, "bnb_4bit_quant_type": "nf4", "bnb_4bit_compute_dtype": torch.bfloat16},
    components_to_quantize="transformer", # instead of ["transformer"]
)

@sayakpaul sayakpaul requested review from SunMarc and yiyixuxu August 25, 2025 12:56
Copy link
Member

@SunMarc SunMarc left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks !

@HuggingFaceDocBuilderDev

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

@stevhliu
Copy link
Member

Will update the docs with this change once merged! 🫡

Copy link
Member

@stevhliu stevhliu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Probably better to put it higher up :)

Copy link
Collaborator

@yiyixuxu yiyixuxu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

thanks!

@yiyixuxu yiyixuxu merged commit eb7ef26 into main Sep 10, 2025
14 checks passed
@yiyixuxu yiyixuxu deleted the allow-non-list-component branch September 10, 2025 19:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants