Description
Bonsai is a 27B model, but takes only about 4GB. Capable of reasoning, coding, vision, and tool use.
Capabilities
Vision Input
Trained for tool use
ReasoningSupports reasoning
Minimum system memory
4GB
Tags
27B
qwen35
Last updated
Updated 2 days agobyREADME
Bonsai 27B
Bonsai 27B is PrismML's family of binary and ternary Qwen3.6 27B models, designed to retain strong reasoning, coding, and agentic capabilities at exceptionally small footprints.
Highlights
- 27B reasoning in a phone-class footprint: binary weights compress the language model to about 3.9 GB, bringing 27B-class local inference to high-end phones.
- Quality that survives extreme compression: across 15 thinking-mode benchmarks, the binary model retains 89.5% of FP16's average score at 1.125 effective bits per weight; the ternary model retains 94.6% at 1.71 bits.
- Ready for demanding local workflows: supports vision input, tool use, hybrid thinking, and a 262K-token context window.
Custom Fields
Special features defined by the model author
Enable Thinking
: boolean
(default=true)
Controls whether the model will think before replying
Preserve Thinking
: boolean
(default=false)
Preserve reasoning content in all prior assistant turns instead of only the most recent one
Parameters
Custom configuration options included with this model
Temperature
0.7
Top K Sampling
20
Top P Sampling
0.95
Sources
The underlying model files this model uses
Based on