FLUX.1-schnell — MLX quant matrix (bf16 / Q8 / Q4)
Pre-quantized, MLX-ready repackagings of black-forest-labs/FLUX.1-schnell
for the SceneWorks native Apple-Silicon worker (mlx-gen).
Each tier is a complete, self-contained turnkey snapshot that loads directly with no in-app
conversion peak (epic 8506). FLUX.1-schnell is a 1–4 step timestep-distilled model (CFG-free).
| Tier | Subdir | Approx. size |
|---|---|---|
| Q4 | q4/ |
~8.7 GB |
| Q8 | q8/ |
~17 GB |
| bf16 | bf16/ |
~31 GB |
All four components are packed in Q4/Q8 — the DiT transformer, the CLIP + T5 text encoders, and the VAE's mid-block attention — using plain asymmetric group-affine quantization (group size 64), byte-identical to the worker's load-time quantization. The bf16 tier is the dense source, mirrored.
License & attribution
Apache License 2.0, inherited from the upstream FLUX.1-schnell release. © Black Forest Labs; this
repository only re-packages the weights (quantization + MLX layout) and adds no new training. See
LICENSE. Original model card: https://huggingface.co/black-forest-labs/FLUX.1-schnell