huggingface / huggingface/diffusers

1 bit, ternary image generation support

Open
#13,925 1 comment 0 reactions 0 assignees View on GitHub
feature-request pipelines quantization
Dominant language
Python
Stars
34.5k
Forks
7.3k
Avg merge
3d 3h
Merged PRs (30d)
91

Description

**Is your feature request related to a problem? Please describe.**
Currently no support for 1 bit image generation

**Describe the solution you'd like.**
[1-bit and Ternary Bonsai Image 4B: Image Generation ](https://prismml.com/news/bonsai-image-4b)

**Describe alternatives you've considered.**
https://github.com/PrismML-Eng/Bonsai-image-demo

Contributor guide

Open the contributing guide

Research direction

Start by reviewing the linked Bonsai Image 4B announcement and demo, then compare their 1-bit and ternary image-generation requirements with diffusers' existing support. Done should be defined as a supported generation path with coverage for both requested modes; the issue names no implementation files or tests.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.