Ternary Bonsai 2 27B Abliterated needs about 9.9 GB of memory at its main file (a 7.2 GB file) with an 8K context window, so it fits on a 12 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.
Ternary Bonsai 2 27B Abliterated – 27B ternary transformer for on‑device inference
Ternary Bonsai 2 27B Abliterated is a 27‑billion‑parameter model quantized to ternary weights. It offers 262,144‑token context and runs on llama.cpp with CUDA or Metal. The model retains 98.2 % of FP16 intelligence while occupying 7.21 GB in PQ2_0 format.
Good at
- Large‑context reasoning
- Fast inference on Apple M5 Max
- Low‑bit ternary weights
Limits
- No vision support
- Requires GPU or CPU with llama.cpp
- Not a final release (preview v0.1)
Best for
- On‑device reasoning tasks
- Developer testing of low‑bit models
- Educational demos of ternary quantization
Who it’s for. Developers and researchers who need a 27B‑class model that fits on a single GPU or laptop. The preview build is suitable for testing and feedback rather than production use.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 7.2 GB | 9.9 GB | Too big | Get file → |
The PQ2_0 pack is 7.21 GB and needs about 9.9 GB of VRAM at 8 k context. For 8 k context on a laptop with 8 GB VRAM, the model will not fit. If you have 12 GB or more, PQ2_0 is the recommended quant. No other quants are available in the file list.
What GPU do you need for Ternary Bonsai 2 27B Abliterated?
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: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is the main file (7.2 GB file, about 9.9 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Ternary Bonsai 2 27B Abliterated
How much VRAM does Ternary Bonsai 2 27B Abliterated need?
About 9.9 GB at its main file with an 8K context window. That is the 7.2 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 2 27B Abliterated on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 9.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Ternary Bonsai 2 27B Abliterated GGUF?
It depends on the quant: file is 7.2 GB.
What context length does Ternary Bonsai 2 27B Abliterated support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.
Where can I download Ternary Bonsai 2 27B Abliterated GGUF?
The files are on Hugging Face in the Hikari07jp/Ternary-Bonsai-2-27B-Abliterated-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 2 27B Abliterated 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 2 27B Abliterated on 8 GB VRAM?
The PQ2_0 pack requires about 9.9 GB of VRAM at 8 k context, so it will not fit on an 8 GB GPU.
Does this model support vision?
No, the vision tower is shipped separately as a Q8_0 mmproj pack.
Is this a final release?
No, it is a preview (v0.1) and not a final release.