ModelListedExternal

Moebius Ft Places2 Mlx Mxfp8

This folder contains a converted MLX version of the ftplaces2 Moebius checkpoint in mxfp8 form. ftplaces2 means the Moebius checkpoint fine-tuned on Places2, for general scene, background, and object inpainting.

Sourcehuggingface.co/iky1e/moebius-ft-places2-mlx-mxfp8/tree/7f0651312e40738c2d11896bab0d854e1c0d7e86

KindUNetBase modelhustvl/MoebiusVersionunet_quantizedBgrade

Model source

Kind: UNet
File size: 412 MB
Base model: hustvl/Moebius
Version: vunet_quantized
Updated: Oct 9, 2026
Source: Hugging Face

Source description

Moebius ft_places2 MLX mxfp8

This folder contains a converted MLX version of the ft_places2 Moebius checkpoint in mxfp8 form. ft_places2 means the Moebius checkpoint fine-tuned on Places2, for general scene, background, and object inpainting.

Original upstream model: hustvl/Moebius
Original source repository: hustvl/Moebius

Moebius checkpoint fine-tuned on Places2. This is the general scene, background, and object inpainting variant converted from the original PyTorch checkpoint.

Identity

Field	Value
Variant name	`ft_places2-mxfp8`
Checkpoint meaning	Fine-tuned on Places2 for general scene, background, and object inpainting.
Original Moebius checkpoint family	`ft_places2`
Original checkpoint type	Places2 fine-tune
Source PyTorch checkpoint	`Moebius-Models/ft_places2/diffusion_pytorch_model.bin`
MLX precision / quantization label	`mxfp8`
Image size	512 x 512
Latent size	64 x 64
Latent channels	4
Mask channels	1
Conditioning IDs	20
VAE scaling factor	0.13025
Noise offset	0.0357

Quantization

MLX MXFP8-style 8-bit export. The manifest selects unet_quantized.safetensors; VAE encoder and decoder remain regular f16 safetensors. Runtime quantization metadata uses mode: mxfp8 with group size 32.

UNet precision mode: mxfp8.
MLX grouped quantization is used for supported linear layers; grouped quantized weights are loaded through the Moebius-MLX manifest/runtime.
Quantization config: 8 bits, group size 32, mxfp8 mode.
The VAE encoder and decoder remain regular f16 safetensors for this variant.

Manifest-selected deployment files

These are the files selected by manifest.json when the Moebius-MLX runtime loads this variant.

Component	File	Size
UNet	`unet_quantized.safetensors`	432.23 MB
VAE encoder	`vae_encoder.safetensors`	68.34 MB
VAE decoder	`vae_decoder.safetensors`	99.00 MB

Files in this folder

unet.safetensors
unet_quantized.safetensors (selected by manifest)
vae_decoder.safetensors (selected by manifest)
vae_encoder.safetensors (selected by manifest)
manifest.json (runtime metadata and file selection)

A minimal runtime package needs manifest.json and the manifest-selected files above. Extra source or fallback files are optional and are not required for inference.

Runtime expectations

This is not a Transformers or Diffusers-native checkpoint. It is intended for the Swift/MLX runtime in Moebius-MLX. The runtime reads manifest.json, loads the selected safetensors files, builds the Moebius UNet and VAE modules, and runs the DDIM inpainting pipeline.

Pipeline constants must match the manifest:

DDIM scheduler: scaled_linear, beta start 0.00085, beta end 0.012, 1000 train timesteps, clip sample false
512 x 512 image resolution and 64 x 64 latent resolution
9-channel UNet input: noisy latent, mask, and masked-image latent
VAE scaling factor 0.13025

Attribution

Moebius was released by the original authors as hustvl/Moebius. This folder is a format conversion and/or quantized MLX packaging of the original PyTorch weights, not a newly trained model.

Requirements

Models

vae_decoder.safetensors
Not resolved
VAE · Hugging Face · iky1e/moebius-ft-places2-mlx-mxfp8
vae_encoder.safetensors
Not resolved
VAE · Hugging Face · iky1e/moebius-ft-places2-mlx-mxfp8

Sources

1 source

Hugging FacePrimaryVersion 7f0651312e40738c2d11896bab0d854e1c0d7e86

huggingface.co/iky1e/moebius-ft-places2-mlx-mxfp8/tree/7f0651312e40738c2d11896bab0d854e1c0d7e86

Verified Oct 9

Model artifacts

1 artifact

unet_quantized.safetensors

safetensors · 412 MB · SHA-256 4e9db583f4a1…93e2 · Hugging Face

Download file

Source excerpts

2 excerpts

ReadmeJul 31huggingface.co/iky1e/moebius-ft-places2-mlx-mxfp8

--- license: apache-2.0 library_name: mlx pipeline_tag: image-to-image base_model: hustvl/Moebius tags: - mlx - image-inpainting - inpainting - diffusion - moebius - ft_places2 - mxfp8 --- # Moebius ft_places2 MLX mxfp8 This folder contains a converted MLX version of the **ft_places2** Moebius checkpoint in **mxfp8** form. `ft_places2` means the Moebius checkpoint fine-tuned on Places2, for general scene, background, and object inpainting. Original upstream model: [hustvl/Moebius](https://huggingface.co/hustvl/Moebius) Original source repository: [hustvl/Moebius](https://github.com/hustvl/Moebius) Moebius checkpoint fine-tuned on Places2. This is the general scene, background, and object inpainting variant converted from the original PyTorch checkpoint. ## Identity | Field | Value | |---|---| | Variant name | `ft_places2-mxfp8` | | Checkpoint meaning | Fine-tuned on Places2 for general scene, background, and object inpainting. | | Original Moebius checkpoint family | `ft_places2` | | Original checkpoint type | Places2 fine-tune | | Source PyTorch checkpoint | `Moebius-Models/ft_places2/diffusion_pytorch_model.bin` | | MLX precision / quantization label | `mxfp8` | | Image size | 512 x 512 | | Latent size | 64 x 64 | | Latent channels | 4 | | Mask channels | 1 | | Conditioning IDs | 20 | | VAE scaling factor | 0.13025 | | Noise offset | 0.0357 | ## Quantization MLX MXFP8-style 8-bit export. The manifest selects `unet_quantized.safetensors`; VAE encoder and decoder remain regular f16 safetensors. Runtime quantization metadata uses `mode: mxfp8` with group size 32. - UNet precision mode: `mxfp8`. - MLX grouped quantization is used for supported linear layers; grouped quantized weights are loaded through the Moebius-MLX manifest/runtime. - Quantization config: 8 bits, group size 32, `mxfp8` mode. - The VAE encoder and decoder remain regular f16 safetensors for this variant. ## Manifest-selected deployment files These are the files selected by `manifest.json` when the Moebius-MLX runtime loads this variant. | Component | File | Size | |---|---|---:| | UNet | `unet_quantized.safetensors` | 432.23 MB | | VAE encoder | `vae_encoder.safetensors` | 68.34 MB | | VAE decoder | `vae_decoder.safetensors` | 99.00 MB | ## Files in this folder - `unet.safetensors` - `unet_quantized.safetensors` (selected by manifest) - `vae_decoder.safetensors` (selec...

NoteJul 31huggingface.co/iky1e/moebius-ft-places2-mlx-mxfp8

Source context: 0 downloads · 0 likes · Pipeline image-to-image · Library mlx · Repo iky1e/moebius-ft-places2-mlx-mxfp8