Ternary Bonsai 1.7B needs about 1.9 GB of memory at Q2_0 (a 0.5 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 4 quants, from 0.5 GB to 3.4 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 3.8 GB.
Ternary Bonsai 1.7B – a 1.7B parameter Qwen3 model in GGUF Q2_0 format
Ternary Bonsai 1.7B is a 1.7‑billion‑parameter Qwen3 model quantized to GGUF Q2_0. It supports a 32,768‑token context window and is released under the Apache‑2.0 license. The model is available in several GGUF files, including a 0.46 GB Q2_0 version and a 3.45 GB F16 version.
Good at
- Compact 1.7B size with 2‑bit ternary quantization
- Supports 32k token context
- Compatible with llama.cpp via Prism fork
Limits
- No vision or multimodal support
- Requires Prism fork for Q2_0 on llama.cpp
- Large memory for F16 variant
Best for
- Low‑memory inference on CPUs
- Rapid prototyping with small VRAM
- Serving lightweight language tasks
Who it’s for. Developers and researchers needing a small, efficient Qwen3 model for local deployment. The model is suitable for edge devices or environments with limited GPU memory.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.5 GB | 1.9 GB | Fits | Get file → | |
| Q2_0 | 0.5 GB | 1.9 GB | Fits | Get file → | |
| Q2_0 | 0.5 GB | 1.9 GB | Fits | Get file → | |
| F16 | 3.4 GB | 4.9 GB | Fits | Get file → |
For 8 GB VRAM, use the 0.46 GB Q2_0 file; it requires about 1.9 GB at 8 k context. If more precision is needed and 12 GB VRAM is available, the 3.45 GB F16 file can be used, estimating 4.9 GB at 8 k context.
What GPU do you need for Ternary Bonsai 1.7B?
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 Q2_0 (0.5 GB file, about 1.9 GB in use).
- 6 GB VRAM: the best quant that fits is F16 (3.4 GB file, about 4.9 GB in use).
Which Ternary Bonsai 1.7B quant should you choose?
Q2_0 (0.5 GB) is the middle option in this repository. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
If memory is tight, the smallest file is the first one in the table at 0.5 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 Ternary Bonsai 1.7B
How much VRAM does Ternary Bonsai 1.7B need?
About 1.9 GB at Q2_0 with an 8K context window. That is the 0.5 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 Ternary Bonsai 1.7B on an 8 GB GPU?
Yes, with F16: the 3.4 GB file needs about 4.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Ternary Bonsai 1.7B GGUF?
It depends on the quant: file is 0.5 GB, Q2_0 is 0.5 GB, Q2_0 is 0.5 GB, F16 is 3.4 GB.
What context length does Ternary Bonsai 1.7B support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.8 GB at 32K.
Where can I download Ternary Bonsai 1.7B GGUF?
The files are on Hugging Face in the prism-ml/Ternary-Bonsai-1.7B-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 Ternary Bonsai 1.7B use?
Hugging Face lists the license as apache-2.0. 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 Ternary Bonsai 1.7B on 8 GB VRAM?
Yes, the 0.46 GB Q2_0 file uses approximately 1.9 GB of VRAM at 8 k context, which fits within 8 GB.
Do I need a special llama.cpp build?
Q2_0 support is not in mainline llama.cpp; use the Prism fork (prism branch) to run the model.
Is the model suitable for vision tasks?
No, the model has no vision or multimodal capabilities.