lmstudio-ai / lmstudio-ai/mlx-engine

[Docs/Question] How to build and use custom mlx-engine with updated mlx-lm for Qwen3-Next-80B-A3B-Instruct-8bit?

Open
#220 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
1.2k
Forks
133
Avg merge
21h 6m
Merged PRs (30d)
1

Description

I tried running `mlx-community/Qwen3-Next-80B-A3B-Instruct-8bit` in LM Studio but the next architecture isn't yet supported in the latest mlx-engine. This is most likely due to the latest mlx-lm release that had a bug.

It works though with the most recent version from GitHub. You can run the model successfully with:

```bash
uv run --with git+https://github.com/ml-explore/mlx-lm.git mlx_lm.chat --model mlx-community/Qwen3-Next-80B-A3B-Instruct-8bit --max-tokens 10000
```

Would it be appropriate to add some docs of how to build all modules - including mlx-engine - from source and using them in LM Studio?

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing the stated mlx-lm command and reviewing how mlx-engine is built and consumed by LM Studio. Document the source-build steps for all required modules, including mlx-engine, and explain how to use those builds in LM Studio. Done means a newcomer can follow the instructions to run the Qwen3-Next model successfully.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
documentation
Issue type
Documentation
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.