MLX models mirrored for Privacy AI
Each top-level folder mirrors one MLX model repository at a pinned commit, so the Privacy AI app can keep offering it for download. Every source file is byte-identical to the source at that commit; each folder adds a MIRROR-NOTES.md with its provenance, and a LICENSE only where the source ships none. Each model keeps its own license, listed below and in its folder. Credit belongs to the model authors and the publishers of the source repositories.
| Folder | Source | Commit | License |
|---|---|---|---|
Bonsai-4B-mlx-1bit/ |
prism-ml/Bonsai-4B-mlx-1bit | 3d25c52f7fa1 |
Apache License 2.0 |
Qwen3.5-0.8B-MLX-4bit/ |
mlx-community/Qwen3.5-0.8B-MLX-4bit | 5d894f8cc4ef |
Apache License 2.0 |
Qwen3.5-2B-MLX-4bit/ |
mlx-community/Qwen3.5-2B-MLX-4bit | 93760be4f1f6 |
Apache License 2.0 |
LFM2.5-1.2B-Instruct-MLX-4bit/ |
LiquidAI/LFM2.5-1.2B-Instruct-MLX-4bit | 7ccafdb04c36 |
LFM Open License v1.0 |
LFM2.5-VL-1.6B-4bit/ |
mlx-community/LFM2.5-VL-1.6B-4bit | 2a53fd7871a8 |
LFM Open License v1.0 |
GLM-OCR-4bit-mlx/ |
EZCon/GLM-OCR-4bit-mlx | 57d7658f94b1 |
MIT License |
gemma-4-e2b-it-4bit/ |
mlx-community/gemma-4-e2b-it-4bit | 238767527555 |
Apache License 2.0 |
Ternary-Bonsai-8B-mlx-2bit/ |
prism-ml/Ternary-Bonsai-8B-mlx-2bit | 9260b24298e4 |
Apache License 2.0 |
Ternary-Bonsai-4B-mlx-2bit/ |
prism-ml/Ternary-Bonsai-4B-mlx-2bit | e1374ad6bf9b |
Apache License 2.0 |
Ternary-Bonsai-1.7B-mlx-2bit/ |
prism-ml/Ternary-Bonsai-1.7B-mlx-2bit | 5f3e306330f6 |
Apache License 2.0 |
granite-4.2-3b-MLX-4bit/ |
lmstudio-community/granite-4.2-3b-MLX-4bit | 4fe06f1699e9 |
Apache License 2.0 |
MiniCPM5-1B-4bit/ |
mlx-community/MiniCPM5-1B-4bit | 36447e84d28c |
Apache License 2.0 |
LFM2.5-2.6B-4bit/ |
mlx-community/LFM2.5-2.6B-4bit | 72a08228cf70 |
LFM Open License v1.0 |
LFM2.5-8B-A1B-MLX-4bit/ |
LiquidAI/LFM2.5-8B-A1B-MLX-4bit | 2e92b640a63d |
LFM Open License v1.0 |
LFM2.5-VL-3B-MLX-4bit/ |
LiquidAI/LFM2.5-VL-3B-MLX-4bit | 7fad79d26092 |
LFM Open License v1.0 |
Ministral-3-3B-Instruct-2512-4bit/ |
mlx-community/Ministral-3-3B-Instruct-2512-4bit | a962dcb09eee |
Apache License 2.0 |
Qwen3.5-4B-MLX-4bit/ |
mlx-community/Qwen3.5-4B-MLX-4bit | 32f3e8ecf654 |
Apache License 2.0 |
Hardware compatibility
Log In to add your hardware
Quantized
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support