LocalClaw/LLM/Bonsai 2 27B
Open-weight local LLM

Bonsai 2 27B

Catalogue summary: PrismML Apache 2.0 ternary Qwen3.8-27B derivative with official GGUF and MLX paths. The PQ2_0 pack is 7.21GB, PTQ1_0 is 5.95GB, and custom llama.cpp/MLX kernels target laptop-class local inference.

Repository editorial metadata; verify comparative claims in the linked upstream material.

16 GB catalogue minimum16 GB RAMPQ2_0Coding assistant
Choose an app

This model needs a special runtime. Unsupported apps are clearly marked.

Compare models
LM StudioNot available for this model
UnslothNot available for this model
OllamaNot available for this model
Open on Hugging FaceFiles, licence and available downloads
llama.cppNot available for this model
Use with LocalClawOptional workspace after the model is installed
This model needs its official runtimePrismML llama.cpp / MLX
Parameters
27.36B (ternary)
Minimum RAM
16 GB
Model size
7.2 GB
Quantization
PQ2_0

Can Bonsai 2 27B run locally?

Bonsai 2 27B has a catalogue minimum of 16 GB RAM with PQ2_0. Actual memory use and speed vary by context length, runtime, backend and system headroom.

Use the official PrismML llama.cpp / MLX setup. The current low-bit files are not a stock LM Studio install.

Model sourcePublic GGUF repository (verified 2026-09-23): prism-ml/Ternary-Bonsai-2-27B-ggufUpstream releasePaper or model cardLicence text
chatcodereasoningvisionagenticmultimodaledgespeed

Install path

01
Check RAM fitMinimum 16 GB RAM. Start with the PQ2_0 quant.
02
Load the modelFollow the official PrismML llama.cpp / MLX instructions. Stock LM Studio support is not confirmed.
03
Control locallyUse LocalClaw to manage models, agents, chat, channels and scheduled OpenClaw work.

Catalogue record

  • Family: bonsai
  • Parameters: 27.36B (ternary)
  • Recommended quantization: PQ2_0
  • Catalogue minimum RAM: 16 GB
  • Catalogue model size: 7.2 GB
  • Tags: chat, code, reasoning, vision, agentic, multimodal, edge, speed, long-context

Practical limits

  • Catalogue RAM is a minimum estimate, not a guarantee for every context length or runtime.
  • Speed and memory use vary by quantization, backend, context length and system headroom.
  • Verify architecture, licence and usage restrictions in the linked upstream material before deployment.

Catalogue tags

  • chat
  • code
  • reasoning
  • vision
  • agentic
  • multimodal

Capability profile

Repository catalogue ratings used by LocalClaw's editorial rubric. They are not a standardized third-party benchmark.

speed
9
quality
9
coding
9
reasoning
9

Technical notes

Developer
PrismML
License
Apache 2.0
Context window
262,144 tokens
Architecture
Qwen3.8-27B-derived hybrid-attention vision-language transformer with end-to-end ternary g128 language weights, custom PTQ1_0/PQ2_0 GGUF packing, optional Q8_0 vision projector and PrismML llama.cpp / MLX kernels.

This model fits these next steps

Hardware fit is based on LocalClaw's RAM tier, model size and quantization metadata. Always leave memory headroom for your OS and runtime.

Starter desktopMac mini M4 16GBPortable fitMacBook Air M4 16GBBest 16GB models16GB RAM guide

Related catalogue entries

Linked mechanically by family, shared tags and nearby RAM tier; this is not a quality ranking.

Bonsai 27B 27.3B (ternary / 1-bit)Agents-A1 4B 4BLing-3.0-tiny 7.9B (1.3B active, MoE)Muse Glimmer 30B 29.8B multimodalNex-N2.5-mini 35B MoEOccamy-1.0 35B (3B active, MoE)Qwen 3.6 35B-A3B 35B (3B active, MoE)Qwen3.8-27B 27B

Where to go next