Hugging Face's logo

moonbox-dev
/
FLUX.1-Fill-dev-MLX

Image-to-Image
MLX
Safetensors
moonbox
flux
inpainting
outpainting
flux-fill
Model card Files
xet
 Community
jennyxue commited on
Commit
070b6ec
·
verified ·
1 Parent(s): 8a1ea16

Model card and the original license

Browse files
Files changed (1) hide show
  1. README.md +19 -0
README.md CHANGED
@@ -48,6 +48,24 @@ for what was measured).
48
  | 8bit | [`8bit/transformer/model.safetensors.index.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/8bit/transformer/model.safetensors.index.json) | 0.1 MB | `a47d31736f65` | reference build: not compared with bf16 (doesn't run in 24 GB); inpaints correctly on the 3 photos |
49
  | 8bit | [`8bit/vae/0.safetensors`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/8bit/vae/0.safetensors) | 165.7 MB | `6ddcda403d28` | reference build: not compared with bf16 (doesn't run in 24 GB); inpaints correctly on the 3 photos |
50
  | 8bit | [`8bit/vae/model.safetensors.index.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/8bit/vae/model.safetensors.index.json) | 17.4 kB | `0a98a344cba2` | reference build: not compared with bf16 (doesn't run in 24 GB); inpaints correctly on the 3 photos |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
51
 
52
  ## How to run
53
 
@@ -76,6 +94,7 @@ seed: `6bit/` and `4bit/` against `8bit/` (nothing bigger fits in 24 GB). Pass:
76
  | Precision | Size | PSNR vs 8bit (mean) | LPIPS (mean / worst) | Status |
77
  |---|---|---|---|---|
78
  | `8bit/` | 17.0 GB | reference | - | not compared with bf16 |
 
79
 
80
  ## How these were made
81
 
 
48
  | 8bit | [`8bit/transformer/model.safetensors.index.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/8bit/transformer/model.safetensors.index.json) | 0.1 MB | `a47d31736f65` | reference build: not compared with bf16 (doesn't run in 24 GB); inpaints correctly on the 3 photos |
49
  | 8bit | [`8bit/vae/0.safetensors`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/8bit/vae/0.safetensors) | 165.7 MB | `6ddcda403d28` | reference build: not compared with bf16 (doesn't run in 24 GB); inpaints correctly on the 3 photos |
50
  | 8bit | [`8bit/vae/model.safetensors.index.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/8bit/vae/model.safetensors.index.json) | 17.4 kB | `0a98a344cba2` | reference build: not compared with bf16 (doesn't run in 24 GB); inpaints correctly on the 3 photos |
51
+ | 6bit | [`6bit/mflux_model.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/mflux_model.json) | 0.4 kB | `9171d42971d1` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
52
+ | 6bit | [`6bit/text_encoder/0.safetensors`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/text_encoder/0.safetensors) | 100.2 MB | `021972a101b1` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
53
+ | 6bit | [`6bit/text_encoder/model.safetensors.index.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/text_encoder/model.safetensors.index.json) | 25.0 kB | `bb76a6a32514` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
54
+ | 6bit | [`6bit/text_encoder_2/0.safetensors`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/text_encoder_2/0.safetensors) | 2.15 GB | `d48839ac576b` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
55
+ | 6bit | [`6bit/text_encoder_2/1.safetensors`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/text_encoder_2/1.safetensors) | 1.72 GB | `4a9a542456a7` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
56
+ | 6bit | [`6bit/text_encoder_2/model.safetensors.index.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/text_encoder_2/model.safetensors.index.json) | 44.1 kB | `ac2d435419f7` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
57
+ | 6bit | [`6bit/tokenizer/tokenizer.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/tokenizer/tokenizer.json) | 3.6 MB | `6d9109cc8389` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
58
+ | 6bit | [`6bit/tokenizer/tokenizer_config.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/tokenizer/tokenizer_config.json) | 0.4 kB | `d0eda7377bb9` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
59
+ | 6bit | [`6bit/tokenizer_2/tokenizer.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/tokenizer_2/tokenizer.json) | 2.4 MB | `2432e8aa8341` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
60
+ | 6bit | [`6bit/tokenizer_2/tokenizer_config.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/tokenizer_2/tokenizer_config.json) | 2.5 kB | `e31ed19bc545` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
61
+ | 6bit | [`6bit/transformer/0.safetensors`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/transformer/0.safetensors) | 2.12 GB | `ff829e699345` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
62
+ | 6bit | [`6bit/transformer/1.safetensors`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/transformer/1.safetensors) | 2.15 GB | `10eb8c736234` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
63
+ | 6bit | [`6bit/transformer/2.safetensors`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/transformer/2.safetensors) | 2.13 GB | `e551955f75c7` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
64
+ | 6bit | [`6bit/transformer/3.safetensors`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/transformer/3.safetensors) | 2.14 GB | `f2c528bcbb50` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
65
+ | 6bit | [`6bit/transformer/4.safetensors`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/transformer/4.safetensors) | 1.14 GB | `c7ee01b73467` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
66
+ | 6bit | [`6bit/transformer/model.safetensors.index.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/transformer/model.safetensors.index.json) | 0.1 MB | `302702931b26` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
67
+ | 6bit | [`6bit/vae/0.safetensors`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/vae/0.safetensors) | 165.2 MB | `5fb147b7776d` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
68
+ | 6bit | [`6bit/vae/model.safetensors.index.json`](https://huggingface.co/moonbox-ai/FLUX.1-Fill-dev-MLX/blob/main/6bit/vae/model.safetensors.index.json) | 17.4 kB | `20fb1f73bf5e` | below threshold (experimental): vs 8bit PSNR 37.8 dB mean, LPIPS 0.056 mean / 0.163 worst |
69
 
70
  ## How to run
71
 
 
94
  | Precision | Size | PSNR vs 8bit (mean) | LPIPS (mean / worst) | Status |
95
  |---|---|---|---|---|
96
  | `8bit/` | 17.0 GB | reference | - | not compared with bf16 |
97
+ | `6bit/` | 13.8 GB | 37.8 dB | 0.056 / 0.163 | **just below threshold, experimental** |
98
 
99
  ## How these were made
100