Instructions to use LocalMuseAI/mlx-flux2-klein-4b-aio-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use LocalMuseAI/mlx-flux2-klein-4b-aio-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download LocalMuseAI/mlx-flux2-klein-4b-aio-4bit --local-dir mlx-flux2-klein-4b-aio-4bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
File size: 1,014 Bytes
cde0259 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 | {
"torch": "2.14.1",
"diffusers": "0.40.0",
"probe": "256x256 reference at RoPE time 10, sigma 1, seed 1729",
"encoderOfficialBF16VsSwiftFP32": {
"relativeL2": 0.004675628049664404,
"cosine": 0.9999891009804717,
"maximumAbsoluteError": 0.0859832763671875
},
"encoderAIOVsSwiftFP32": {
"relativeL2": 2.581435821710404e-06,
"cosine": 0.9999999999966654,
"maximumAbsoluteError": 2.9742717742919922e-05
},
"referenceTokensVsSwift": {
"relativeL2": 2.5848853115520615e-06,
"cosine": 0.9999999999966559,
"maximumAbsoluteError": 1.6480684280395508e-05
},
"editBF16VsSwiftInt4": {
"relativeL2": 0.09492114942412691,
"cosine": 0.9954871472526797,
"maximumAbsoluteError": 1.145721435546875
},
"editDequantizedVsSwiftInt4": {
"relativeL2": 0.010690767753481906,
"cosine": 0.9999429995098634,
"maximumAbsoluteError": 0.125
},
"editVsTextOnlySwiftRelativeL2": 0.8220424651367878,
"elapsedSeconds": 134.72175112500008,
"passed": true
}
|