| --- |
| language: |
| - en |
| license: other |
| license_name: flux-non-commercial-license |
| license_link: LICENSE.md |
| tags: |
| - image-generation |
| - image-editing |
| - flux |
| - mlx |
| - apple-silicon |
| pipeline_tag: text-to-image |
| library_name: mlx |
| base_model: black-forest-labs/FLUX.2-klein-9B |
| --- |
| |
| # FLUX.2 [klein] 9B — SceneWorks MLX turnkey |
|
|
| A public, ungated re-host of [black-forest-labs/FLUX.2-klein-9B](https://huggingface.co/black-forest-labs/FLUX.2-klein-9B), |
| pre-quantized for Apple Silicon and consumed natively by [SceneWorks](https://github.com/SceneWorks). |
| No Hugging Face token or license click is needed before downloading — the FLUX Non-Commercial |
| License (`LICENSE.md`) travels with the weights, and generations remain non-commercial use only. |
|
|
| ## Tiers |
|
|
| Each subdirectory is a complete, ready-to-run MLX tree (transformer + Qwen3 text encoder + VAE + |
| tokenizer + `model_index.json`). Only the transformer is quantized — the 8B Qwen3 text encoder stays |
| **full-precision bf16 in every tier** for fidelity. Pick one based on your Mac's unified memory: |
|
|
| | Tier | Path | Transformer | Text encoder | Approx. size | |
| |------|------|-------------|--------------|--------------| |
| | Q4 (default) | `q4/` | packed 4-bit | dense bf16 | ~22 GB | |
| | Q8 | `q8/` | packed 8-bit | dense bf16 | ~26 GB | |
| | bf16 | `bf16/` | dense bf16 | dense bf16 | ~35 GB | |
|
|
| The packed transformer self-describes its quantization, so the loader reads each tier as-is |
| (no load-time conversion). |
|
|
| ## License |
|
|
| FLUX Non-Commercial License — see [`LICENSE.md`](LICENSE.md) and the |
| [Acceptable Use Policy](https://bfl.ai/legal/usage-policy). Non-commercial use only. |
|
|
