lmstudio-ai / lmstudio-ai/mlx-engine

Support for 1bit models

Open
#299 5 comments 57 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
1.2k
Forks
133
Avg merge
21h 6m
Merged PRs (30d)
1

Description

I just downloaded https://huggingface.co/prism-ml/Bonsai-8B-mlx-1bit and got

`Error when loading model: ValueError: [quantize] The requested number of bits 1 is not supported. The supported bits are 2, 3, 4, 5, 6 and 8.
`

I'm on macos, arm, lm studio version 0.4.8

Contributor guide

Open the contributing guide

Research direction

Start by reproducing the loading error with the Bonsai-8B-mlx-1bit model on macOS ARM using LM Studio 0.4.8. Trace the model-loading path where the requested quantization bits are validated, then verify that the model loads successfully and that existing 2-, 3-, 4-, 5-, 6- and 8-bit models still work.

Written by the indexing model from the issue text.

Assessment

Tech stack
macos, python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
52/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.