What are the hardware requirements for Bonsai 27B MLX BF16 Config Repaired?

Bonsai 27B MLX BF16 Config Repaired requires approximately 53.5 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Bonsai 27B MLX BF16 Config Repaired and how large is it?

Bonsai 27B MLX BF16 Config Repaired is developed by TiGa-RCE, a model with 27b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Bonsai 27B MLX BF16 Config Repaired?

Bonsai 27B MLX BF16 Config Repaired is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Bonsai 27B MLX BF16 Config Repaired by TiGa-RCE — VRAM 53.5GB

Name: Bonsai 27B MLX BF16 Config Repaired
Author: TiGa-RCE

Bonsai 27B MLX BF16 Config Repaired is an open-source language model by TiGa-RCE. Features: 27b LLM, VRAM: 53.5GB, License: apache-2.0, Quantized, LLM Explorer Score: 0.29.

Apple-silicon Base model:finetune:prism-ml/b... Base model:prism-ml/bonsai-27b... Bonsai Conversational En Experimental Gguf Mlx Quantized Qwen3.6 Qwen3 5 Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/TiGa-RCE/Bonsai-27B-MLX-BF16-Config-Repaired

Bonsai 27B MLX BF16 Config Repaired Parameters and Internals

LLM Name	Bonsai 27B MLX BF16 Config Repaired
Repository 🤗	https://huggingface.co/TiGa-RCE/Bonsai-27B-MLX-BF16-Config-Repaired
Base Model(s)	prism-ml/Bonsai-27B-gguf prism-ml/Bonsai-27B-gguf
Model Size	27b
Required VRAM	53.5 GB
Updated	2026-08-08
Maintainer	TiGa-RCE
Model Type	qwen3_5
Model Files	5.2 GB: 1-of-11 5.3 GB: 2-of-11 5.3 GB: 3-of-11 5.3 GB: 4-of-11 5.3 GB: 5-of-11 5.3 GB: 6-of-11 5.3 GB: 7-of-11 5.3 GB: 8-of-11 5.3 GB: 9-of-11 3.4 GB: 10-of-11 2.5 GB: 11-of-11
Supported Languages	en
GGUF Quantization	Yes
Quantization Type	gguf
Model Architecture	Qwen3_5ForConditionalGeneration
License	apache-2.0
Model Max Length	262144
Transformers Version	4.57.1
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Bonsai 27B MLX BF16 Config Repaired capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Bonsai 27B MLX BF16 Config Repaired

Best Alternatives	Context / RAM	Downloads	Likes
Qwen3.8 27B GGUF	0K / 0.9 GB	0	685
... Opus Reasoning Distilled GGUF	0K / 0.9 GB	67921	679
...en3.8 27B PristinelyUncensored	0K / 55.5 GB	2244	5
Qwen3.8 27B MIX GGUF	0K / 0.9 GB	141	3
... Opus Reasoning Distilled GGUF	0K / 10.7 GB	9907	48
Qwen3.6 27B OptiQ 4bit	256K / 18.7 GB	13065	59
Bonsai 27B Mlx 1bit	0K / 5.1 GB	809309	205
Qwen3.8 27B 8bit	0K / 29.4 GB	13892	33
Ternary Bonsai 27B Mlx 2bit	0K / 8.5 GB	809024	168
Qwen3.8 27B 4bit	0K / 16 GB	6055	27

Note: green Score (e.g. "73.2") means that the model is better than TiGa-RCE/Bonsai-27B-MLX-BF16-Config-Repaired.