Skip to content

Conversation

@JyotinderSingh
Copy link
Collaborator

No description provided.

@gemini-code-assist
Copy link
Contributor

Summary of Changes

Hello @JyotinderSingh, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request introduces comprehensive documentation for 4-bit integer quantization (INT4) within the Keras framework. The new guide explains the principles, benefits, and practical application of INT4 quantization, including code examples for both custom and KerasHub models. It also provides detailed performance benchmarks across various model types and offers guidance on its appropriate use and limitations, thereby enhancing Keras's capabilities for efficient model deployment.

Highlights

  • New INT4 Quantization Guide: A comprehensive guide for 4-bit integer quantization (INT4) in Keras has been added, detailing its principles, benefits, and practical application.
  • Practical Examples: The guide includes code examples demonstrating how to quantize both minimal functional models and KerasHub models (e.g., Gemma3), along with saving and reloading quantized models.
  • Performance Benchmarks: Detailed micro-benchmarks are provided for various models (DistilBERT, Falcon, Gemma3, Llama 3.2, OPT), showcasing the impact of INT4 quantization on accuracy, model size, GPU memory, latency, and throughput.
  • Usage Guidance and Limitations: The documentation offers clear advice on when to prefer INT4 over INT8, practical tips for implementation, and outlines the current limitations of INT4 quantization in Keras.
  • Documentation Integration: The new INT4 quantization guide has been integrated into the main Keras documentation structure, making it accessible to users.
Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature Command Description
Code Review /gemini review Performs a code review for the current pull request in its current state.
Pull Request Summary /gemini summary Provides a summary of the current pull request in its current state.
Comment @gemini-code-assist Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help /gemini help Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

  1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution.

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

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

Code Review

This pull request adds excellent and thorough documentation for INT4 quantization in Keras. The guide is well-written, covering the topic from theory to practice with useful code examples and performance benchmarks. This is a great contribution to the Keras documentation. I have only one minor suggestion for a grammatical correction to enhance clarity.


### Text Classification (DistilBERT Base on SST-2)

<img class="k-inline-icon" src="https://colab.research.google.com/img/colab_favicon.ico"/> [**View in Colab**](https://colab.research.google.com/gist/JyotinderSingh/77e874187d6da3f8280c053192f78c06/int4-quantization-micro-benchmark-distilbert.ipynb)
Copy link
Contributor

Choose a reason for hiding this comment

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

Are these links going stay in the long run?

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

I'll keep them live, but I'm going to replace the links with a permanent version as soon as I figure out the best long-term solution.

@hertschuh
Copy link
Contributor

@JyotinderSingh Can you rebase?

I don't know which guide you want first in the menu, the int8 one or the int4 one?

JyotinderSingh and others added 4 commits October 16, 2025 06:39
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
@JyotinderSingh
Copy link
Collaborator Author

@JyotinderSingh Can you rebase?

I don't know which guide you want first in the menu, the int8 one or the int4 one?

Resolved the conflicts

@hertschuh hertschuh merged commit ade0c30 into keras-team:master Oct 16, 2025
3 checks passed
@JyotinderSingh JyotinderSingh deleted the int4-docs branch October 16, 2025 05:24
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.

3 participants