snack models needs about 3.2 GB of memory at its main file (a 1.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 6 quants, from 1.0 GB to 5.5 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 6.4 GB.
Snack Models – 1.73B parameter LFM3 model with reasoning, tools, and vision abilities
Snack Models is a 1.73B parameter LFM3 architecture designed for reasoning, tool use, and vision tasks. It supports a large context window of 128000 tokens and is available in multiple GGUF quantizations. The model is released under an other license.
Good at
- Reasoning
- Tool usage
- Vision tasks
Limits
- Large context requires high VRAM
- No explicit vision support in quant files
- License not specified
Best for
- Complex reasoning queries
- Tool‑integrated workflows
- Vision‑enabled prompts
Who it’s for. Developers and researchers needing a versatile 1.73B LFM3 model with reasoning, tool, and vision capabilities. Suitable for applications that require a large context window and multi‑modal support.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 1.0 GB | 3.2 GB | Fits | Get file → | |
| — | 1.0 GB | 3.2 GB | Fits | Get file → | |
| — | 1.0 GB | 3.2 GB | Fits | Get file → | |
| — | 1.0 GB | 3.2 GB | Fits | Get file → | |
| — | 2.6 GB | 4.7 GB | Fits | Get file → | |
| — | 5.5 GB | 7.6 GB | Tight | Get file → |
For 8‑GB VRAM, use the 1.05‑GB GGUF file which estimates 3.2 GB memory at 8k context. If 12‑GB VRAM is available, the 2.61‑GB file uses about 4.7 GB memory. For 16‑GB VRAM or higher, the 5.45‑GB file requires roughly 7.6 GB memory. Estimates are approximate.
What GPU do you need for snack models?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is the main file (1.0 GB file, about 3.2 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (2.6 GB file, about 4.7 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (5.5 GB file, about 7.6 GB in use).
Which snack models quant should you choose?
If memory is tight, the smallest file is the first one in the table at 1.0 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about snack models
How much VRAM does snack models need?
About 3.2 GB at its main file with an 8K context window. That is the 1.0 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run snack models on an 8 GB GPU?
Yes, with the smaller files: the 5.5 GB file needs about 7.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of snack models GGUF?
It depends on the quant: file is 1.0 GB, file is 1.0 GB, file is 1.0 GB, file is 1.0 GB, file is 2.6 GB, file is 5.5 GB.
What context length does snack models support?
The GGUF header lists a trained context length of 128,000 tokens (125K). The KV cache grows with the context you use: about 1.6 GB at 8K and 6.4 GB at 32K.
Where can I download snack models GGUF?
The files are on Hugging Face in the TurtletortOS/snack-models repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does snack models use?
Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Snack Models on 8 GB VRAM?
Yes, the 1.05‑GB GGUF file should fit in 8 GB VRAM with an estimated 3.2 GB memory usage at 8k context.
Does the model support vision tasks?
Yes, the model has detected vision abilities, but the quant files do not specify explicit vision support.