Goekdeniz-Guelmez / MLX-LoRA-Studio
A native Mac App for LLM fine-tuning on Apple Silicon — fully on-device, fully open source.
Why this job: says: lora; topics: rlhf, llm-training, reinforcement-learning, fine tuning
What it does
- 12 training algorithms out of the box. SFT, DPO, DSLA, FTPO, CPO, ORPO, GRPO, KLPO,
- Selectable SFT losses. NLL, Chunked NLL for lower peak memory, and DFT for emphasizing difficult tokens.
- LoRA, DoRA, QLoRA (4/6/8-bit), full fine-tuning, and QAT (Quantization-Aware Training
- Adapter resume. point a new run at an existing adapter checkpoint and Studio picks
- Judge / reward model selection for RL-style algorithms (GRPO, XPO, Online DPO, PPO,
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Active
- Last commit 13 days ago
- Stars
- 287
- +3 this week
- Licence
- MIT
- Allows business use
- Downloads for
- Mac
- From the latest release
- Latest version
- v3.0.0
- Released 13 days ago
- Built with
- Swift
- 83% of the code
Is it looked after?
- Contributors
- 1
- Last 90 days
- Issues
- 100% closed
- 0 still open
- Pull requests
- 28 merged
- 0 waiting
- Releases
- 10
- 10 in the last year
- No contributing guide
- No security policy
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| hiyouga / EasyR1 | 5,211 | yesterday | Apache-2.0 | No downloads |
| Human-Agent-Society / reef | 7,787 | today | Apache-2.0 | No downloads |
| oumi-ai / oumi | 9,429 | 3 days ago | Apache-2.0 | No downloads |
More: download files, languages, topics and project record
Languages
- Swift 83%
- Python 15%
- Shell 2%
Topics
- deep-learning
- llm-training
- llms
- machine-learning
- mlx
- mlx-lm
- mlx-lm-lora
- nlp
- preference-learning
- reinforcement-learning
- rl
- rlhf
Project record
- JobFine-tuning
- StatusListed
- Commits157
- Forks31
- Watching3
- Stars in 28 days+11
- CreatedJun 11, 2026
- First seen by usNov 7, 2026
- CheckedNov 7, 2026