TE

Bonsai 2 27B PTQ1 0 MTP Uncensored Ready

TernaryGGUF by ayshinkoUpdated 2026-09-22
Download on Hugging Face
Parameters2M
Context—
Layers—
Architectureqwen35
Smallest quant0.0 GB
Quick answer

Bonsai 2 27B PTQ1 0 MTP Uncensored Ready needs about 0.5 GB of memory at its main file (a 0.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 0.0 GB to 6.4 GB (Q8_0).

27B-class local LLM with MTP speculative decoding

Bonsai 2 27B PTQ1_0 MTP is a 27B-class model built for a single consumer GPU with roughly 12 GB VRAM. It uses PrismML’s PTQ1_0 ternary quantization and ProCreations’ MTP head exported with a Q8_0 prediction head.

Good at

  • Fast short‑context inference
  • Multi‑token prediction
  • Uncensored‑ready with optional LoRA

Limits

  • No vision support
  • Requires CUDA‑optimized llama.cpp
  • License is apache‑2.0

Best for

  • Local chatbot deployments
  • Rapid prototyping on 12 GB GPUs
  • Testing speculative decoding

Who it’s for. Developers and researchers who need a 27B‑class model that runs on a single consumer GPU. The model is suitable for local inference without external services and supports optional uncensored behavior via a separate LoRA.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 0.5 GB
Fits Get file →
Q8_06.4 GB 7.7 GB
Tight Get file →

For 8 GB VRAM use the unknown quant file (0.01 GB, ~0.5 GB at 8k context). For 12 GB VRAM use the Q8_0 quant file (6.4 GB, ~7.7 GB at 8k context). Memory estimates are approximate.

What GPU do you need for Bonsai 2 27B PTQ1 0 MTP Uncensored Ready?

These figures add the file size, an fp16 KV cache at an 8K context window 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 (0.0 GB file, about 0.5 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 0.5 GB in use).
  • 8 GB VRAM: the best quant that fits is Q8_0 (6.4 GB file, about 7.7 GB in use).

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters2M

Source

Quantized byayshinko
FamilyTernary
Licenseapache-2.0
Downloads (HF)1,378
Last updated2026-09-22

Frequently asked questions about Bonsai 2 27B PTQ1 0 MTP Uncensored Ready

How much VRAM does Bonsai 2 27B PTQ1 0 MTP Uncensored Ready need?

About 0.5 GB at its main file 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 Bonsai 2 27B PTQ1 0 MTP Uncensored Ready on an 8 GB GPU?

Yes, with Q8_0: the 6.4 GB file needs about 7.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Bonsai 2 27B PTQ1 0 MTP Uncensored Ready GGUF?

It depends on the quant: file is 0.0 GB, Q8_0 is 6.4 GB.

Where can I download Bonsai 2 27B PTQ1 0 MTP Uncensored Ready GGUF?

The files are on Hugging Face in the ayshinko/Bonsai-2-27B-PTQ1_0-MTP-Uncensored-Ready-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 Bonsai 2 27B PTQ1 0 MTP Uncensored Ready 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 this model on 8 GB VRAM?

Yes, use the unknown quant file. It is 0.01 GB and requires about 0.5 GB at 8k context, well within 8 GB VRAM.

What license does this model use?

The model is released under the apache‑2.0 license.

Is the model uncensored by default?

The shipped GGUF is uncensored‑ready but uncensored behavior is off by default; an optional OrcaRouter LoRA can enable it.

Other GGUFs of Ternary-Bonsai-2-27B-MTP

Similar models

Browse all →

More from ayshinko

All by ayshinko →

More Ternary models

All Ternary →

Other tiny models

All tiny models →