Lantern: offline private AI
Collection
Models for Lantern, an AI app that runs entirely on your iPhone, iPad or Mac. No account, no server, nothing leaves the device. • 7 items • Updated
SmolVLM 500M Instruct, 4-bit MLX, for Lantern
A copy of mlx-community/SmolVLM-500M-Instruct-4bit, kept here for Lantern, an offline, private AI app for iPhone, iPad and Mac. The model runs entirely on the device: no account, no server, and nothing you type leaves it. The only network use is downloading these files once.
The weights are unchanged from the mlx-community conversion. The original model is HuggingFaceTB/SmolVLM-500M-Instruct.
| Parameters | 0.5 billion |
| Precision | 4-bit, MLX format |
| Download | about 0.3 GB |
| Takes in | text and photos |
| Runs on | Any iPhone Lantern supports. |
Good for: Looks at a photo and describes it. Tiny, so it's quick, but it misses detail.
Use it in Lantern
Lantern already lists this model. To download this copy instead, open Settings, turn on Show advanced options, choose Add a model from Hugging Face and enter RabiatS/SmolVLM-500M-Instruct-4bit-Lantern. Lantern checks it against your device's memory before anything downloads.
Use it with mlx-lm on a Mac
pip install mlx-vlm
mlx_vlm.generate --model RabiatS/SmolVLM-500M-Instruct-4bit-Lantern --image photo.jpg --prompt "What is in this picture?"
Licence
Apache 2.0, as published with HuggingFaceTB/SmolVLM-500M-Instruct.
Credit for the model goes to its authors, and for the MLX conversion to mlx-community.
- Downloads last month
- 27
Model size
0.5B params
Tensor type
U32
·
F16 ·
Hardware compatibility
Log In to add your hardware
Quantized
Model tree for RabiatS/SmolVLM-500M-Instruct-4bit-Lantern
Base model
HuggingFaceTB/SmolLM2-360M Quantized
HuggingFaceTB/SmolLM2-360M-Instruct Quantized
HuggingFaceTB/SmolVLM-500M-Instruct