tinygemma3 needs about 0.7 GB of memory at Q8_0 (a 0.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 0.5 GB.
tinygemma3 – a 38‑million‑parameter vision model in GGUF format
tinygemma3 is a 0.038 B parameter gemma3 model trained on CIFAR‑10. It supports vision tasks and is available in a single Q8_0 quantized GGUF file. The model is released under the WTFPL license.
Good at
- Vision inference on small images
- Fast inference with Q8_0 quant
- Low‑memory usage
Limits
- No text‑only support
- Requires 8 k context for full operation
- Only Q8_0 quant available
Best for
- Image classification on embedded devices
- Rapid vision prototyping
- Testing vision pipelines in llama.cpp
Who it’s for. Developers needing a lightweight vision model for quick experimentation. The model is suitable for educational projects and CI testing of vision capabilities in llama.cpp.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q8_0 | 0.0 GB | 0.7 GB | Fits | Get file → |
The Q8_0 file is 0.05 GB and needs about 0.7 GB of VRAM at 8 k context. This is the only quant available, so use it for any deployment that can accommodate the memory estimate.
What GPU do you need for tinygemma3?
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 Q8_0 (0.0 GB file, about 0.7 GB in use).
Which tinygemma3 quant should you choose?
Q8_0 (0.0 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about tinygemma3
How much VRAM does tinygemma3 need?
About 0.7 GB at Q8_0 with an 8K context window. That is the 0.0 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 tinygemma3 on an 8 GB GPU?
Yes, with Q8_0: the 0.0 GB file needs about 0.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of tinygemma3 GGUF?
It depends on the quant: Q8_0 is 0.0 GB.
What context length does tinygemma3 support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.1 GB at 8K and 0.5 GB at 32K.
Where can I download tinygemma3 GGUF?
The files are on Hugging Face in the ggml-org/tinygemma3-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does tinygemma3 use?
Hugging Face lists the license as wtfpl. 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 tinygemma3 on 8 GB VRAM?
Yes, the Q8_0 quant requires roughly 0.7 GB of VRAM at 8 k context, well within an 8 GB budget.
Does tinygemma3 support text generation?
No, the model is trained for vision tasks and does not provide text generation capabilities.
What license applies to tinygemma3?
The model is released under the WTFPL license, allowing free use and modification.