# PrismML: Ternary Bonsai 2 27B

Provider: prism-ml  
Category: llm  
Model ID: `prism-ml/ternary-bonsai-2-27b`

Bonsai 2 27B is a 27B-parameter reasoning model from PrismML derived from Qwen3.8-27B. It supports coding, mathematics, tool calling, and image understanding with a 262K-token context window. Ternary compression shrinks...

## Specs

- Context length: 262144 tokens
- Max output: 32768 tokens
- Modalities: text, image
- Released: 2026-09-18

## Pricing

- Input: $0.07 per million tokens
- Output: $0.50 per million tokens

## Providers

- **prism-ml** — ctx 262144, input $0.07/M, output $0.50/M

## Verdict

PrismML's Ternary Bonsai 2 27B delivers a 262K context window at aggressive pricing — $0.07 input makes it one of the cheapest long-context options available. The ternary quantization architecture keeps inference costs low while maintaining reasonable quality for document-heavy workflows. Without public benchmarks, you're trading proven performance data for cost savings. Reach for this when budget constraints matter more than leaderboard rankings, especially for high-volume document processing where input tokens dominate your bill.

---
Last verified: 2026-10-19T02:30:34.239Z  
Canonical URL: https://switchy.build/directory/models/ternary-bonsai-2-27b