Text-to-Image
MLX
MiniMax H3
Safetensors
English
apple-silicon
macos
image-to-image
image-editing
image-generation
Instructions to use devin-lai/MiniMax-H3-Image-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use devin-lai/MiniMax-H3-Image-mlx with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download devin-lai/MiniMax-H3-Image-mlx --local-dir MiniMax-H3-Image-mlx
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download comparisons/optimization-v2/decoder-portrait/comparison/metrics.json from devin-lai/MiniMax-H3-Image-mlx: direct link, hf CLI and curl.
- Browser
- Download file 4.48 kB
-
https://huggingface.co/devin-lai/MiniMax-H3-Image-mlx/resolve/main/comparisons/optimization-v2/decoder-portrait/comparison/metrics.json
- Command line
-
hf download hf://devin-lai/MiniMax-H3-Image-mlx/comparisons/optimization-v2/decoder-portrait/comparison/metrics.json
-
curl -L -o metrics.json https://huggingface.co/devin-lai/MiniMax-H3-Image-mlx/resolve/main/comparisons/optimization-v2/decoder-portrait/comparison/metrics.json
4.48 kB
| { | |
| "validation_scope": "Same final latent; decoder only", | |
| "reference": "/Users/devin/Desktop/MiniMax-H3-Image-mlx/comparisons/optimization-v2/decoder-portrait/native-b1.png", | |
| "native_size": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "crop_xyxy": [ | |
| 768, | |
| 768, | |
| 1280, | |
| 1280 | |
| ], | |
| "heatmap_gain": 8, | |
| "metric_definition": { | |
| "psnr": "Full-resolution 8-bit RGB MSE, data range 255; null means infinity", | |
| "ssim": "scikit-image RGB structural_similarity, channel_axis=-1, data_range=255, default 7x7 window", | |
| "mae": "Full-resolution mean absolute 8-bit RGB error" | |
| }, | |
| "rows": [ | |
| { | |
| "label": "native-b1", | |
| "image": "/Users/devin/Desktop/MiniMax-H3-Image-mlx/comparisons/optimization-v2/decoder-portrait/native-b1.png", | |
| "png_sha256": "32af5e3b95919e79e0019058ec396824c5861f9a43a4fa50c7b20348e8d81933", | |
| "pixel_identical": true, | |
| "mae_8bit": 0.0, | |
| "max_error_8bit": 0, | |
| "psnr_db": null, | |
| "ssim_rgb": 1.0, | |
| "timings": { | |
| "decode_s": 5.739497271017171 | |
| }, | |
| "steps": [ | |
| 0, | |
| 1, | |
| 2, | |
| 3, | |
| 4, | |
| 5 | |
| ], | |
| "timing_scope": "Warm decoder only; model loading excluded" | |
| }, | |
| { | |
| "label": "native-b4", | |
| "image": "/Users/devin/Desktop/MiniMax-H3-Image-mlx/comparisons/optimization-v2/decoder-portrait/native-b4.png", | |
| "png_sha256": "32af5e3b95919e79e0019058ec396824c5861f9a43a4fa50c7b20348e8d81933", | |
| "pixel_identical": true, | |
| "mae_8bit": 0.0, | |
| "max_error_8bit": 0, | |
| "psnr_db": null, | |
| "ssim_rgb": 1.0, | |
| "timings": { | |
| "decode_s": 2.6730494999792427 | |
| }, | |
| "steps": [ | |
| 0, | |
| 1, | |
| 2, | |
| 3, | |
| 4, | |
| 5 | |
| ], | |
| "timing_scope": "Warm decoder only; model loading excluded" | |
| }, | |
| { | |
| "label": "compiled-b4", | |
| "image": "/Users/devin/Desktop/MiniMax-H3-Image-mlx/comparisons/optimization-v2/decoder-portrait/compiled-b4.png", | |
| "png_sha256": "32af5e3b95919e79e0019058ec396824c5861f9a43a4fa50c7b20348e8d81933", | |
| "pixel_identical": true, | |
| "mae_8bit": 0.0, | |
| "max_error_8bit": 0, | |
| "psnr_db": null, | |
| "ssim_rgb": 1.0, | |
| "timings": { | |
| "decode_s": 1.929837999981828 | |
| }, | |
| "steps": [ | |
| 0, | |
| 1, | |
| 2, | |
| 3, | |
| 4, | |
| 5 | |
| ], | |
| "timing_scope": "Warm decoder only; model loading excluded" | |
| }, | |
| { | |
| "label": "native-b8", | |
| "image": "/Users/devin/Desktop/MiniMax-H3-Image-mlx/comparisons/optimization-v2/decoder-portrait/native-b8.png", | |
| "png_sha256": "a34540dbc2f0f2249df4799b71892d980923bdc972e8d4db5de7e9a9ab08b88a", | |
| "pixel_identical": false, | |
| "mae_8bit": 0.15126427014668783, | |
| "max_error_8bit": 5, | |
| "psnr_db": 55.96620402076142, | |
| "ssim_rgb": 0.9991664658666948, | |
| "timings": { | |
| "decode_s": 2.2933986249845475 | |
| }, | |
| "steps": [ | |
| 0, | |
| 1, | |
| 2, | |
| 3, | |
| 4, | |
| 5 | |
| ], | |
| "timing_scope": "Warm decoder only; model loading excluded" | |
| }, | |
| { | |
| "label": "compiled-b8", | |
| "image": "/Users/devin/Desktop/MiniMax-H3-Image-mlx/comparisons/optimization-v2/decoder-portrait/compiled-b8.png", | |
| "png_sha256": "a34540dbc2f0f2249df4799b71892d980923bdc972e8d4db5de7e9a9ab08b88a", | |
| "pixel_identical": false, | |
| "mae_8bit": 0.15126427014668783, | |
| "max_error_8bit": 5, | |
| "psnr_db": 55.96620402076142, | |
| "ssim_rgb": 0.9991664658666948, | |
| "timings": { | |
| "decode_s": 1.6636835205135867 | |
| }, | |
| "steps": [ | |
| 0, | |
| 1, | |
| 2, | |
| 3, | |
| 4, | |
| 5 | |
| ], | |
| "timing_scope": "Warm decoder only; model loading excluded" | |
| }, | |
| { | |
| "label": "coreml-CPU_AND_GPU", | |
| "image": "/Users/devin/Desktop/MiniMax-H3-Image-mlx/comparisons/optimization-v2/decoder-portrait/coreml-CPU_AND_GPU.png", | |
| "png_sha256": "ad830fa42d204982381787317eee53f4c662d512cbafe260a53e065b987c18c6", | |
| "pixel_identical": false, | |
| "mae_8bit": 0.2621665795644124, | |
| "max_error_8bit": 11, | |
| "psnr_db": 53.128439249639484, | |
| "ssim_rgb": 0.998681981765892, | |
| "timings": { | |
| "decode_s": 1.591182250645943 | |
| }, | |
| "steps": [ | |
| 0, | |
| 1, | |
| 2, | |
| 3, | |
| 4, | |
| 5 | |
| ], | |
| "timing_scope": "Warm decoder only; model loading excluded" | |
| } | |
| ], | |
| "interpretation": "Pixel similarity to the same seeded trajectory, not a measure of prompt adherence or perceptual quality." | |
| } | |