Skip to content

[docs] quant_kwargs#11712

Merged
stevhliu merged 2 commits into
huggingface:mainfrom
stevhliu:quant-kwargs
Jul 29, 2025
Merged

[docs] quant_kwargs#11712
stevhliu merged 2 commits into
huggingface:mainfrom
stevhliu:quant-kwargs

Conversation

@stevhliu

Copy link
Copy Markdown
Member

Improves where you can find the quant_kwargs in the PipelineQuantizationConfig docs

@stevhliu stevhliu requested review from asomoza and sayakpaul June 14, 2025 00:40

@sayakpaul sayakpaul left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

Thanks! Do we have to include an explicit point on how users can derive qaunt_kwargs for a given backend?

@asomoza asomoza left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

I'm undecided as to what is best, to include each quantization backend here as @sayakpaul mentioned or to an index of them like you did, so I'll leave it to you.

Also I wanted to see the final docs but the doc builder failed.

@stevhliu

stevhliu commented Jun 16, 2025

Copy link
Copy Markdown
Member Author

I think it's probably best to refer the user to the specific API docs for each backend where they can find all the relevant arguments, and keep this one more high-level.

Not sure why the doc-builder is failing though. bitsandbytes should already be installed in the CI test environment no? 🤔

@sayakpaul

Copy link
Copy Markdown
Member

I think it's probably best to refer the user to the specific API docs for each backend where they can find all the relevant arguments, and keep this one more high-level.

Yeah indeed. I was asking for an explicit note that says just that.

@sayakpaul

Copy link
Copy Markdown
Member

Not sure why the doc-builder is failing though. bitsandbytes should already be installed in the CI test environment no? 🤔

No actually. bitsandbytes is not a required dependency in our CI and we install it only when running the quantization tests.

I opened #11728 for this. Let's see.

@HuggingFaceDocBuilderDev

Copy link
Copy Markdown

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 stevhliu merged commit dfa4883 into huggingface:main Jul 29, 2025
1 check passed
@stevhliu stevhliu deleted the quant-kwargs branch July 29, 2025 17:23
- for granular quantization control, define a `quant_mapping` that provides the quantization configuration for individual model components

### Simple quantization
### Basic quantization

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

in line 94 there's a link to 'simple quantization' so this will break it, change the other link then?

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

saw this too late I think

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

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

Oops, should be fixed in #12018!

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.

4 participants