---
id: "prism-ml/ternary-bonsai-2-27b"
also: ["prism-ml/ternary-bonsai-2-27b-20260918", "prism-ml/Ternary-Bonsai-2-27B-gguf"]
cheapest_list_usd_per_m: "0.075 in, 0.5 out, at darkbloom/int4 on openrouter"
cheapest_effective_usd_per_m: "0.0724 in, 0.4995 out, at darkbloom/int4 on openrouter"
context: 262144
max_output: 32768
input: ["text", "image"]
output: ["text"]
released: "2026-09-18"
reasoning: ["xhigh", "medium", "none"]
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "include_reasoning", "logprobs", "max_tokens", "presence_penalty", "reasoning", "reasoning_effort", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 1
gateways: "openrouter (1)"
no_training: 1
trains: 0
retains_prompts: 1
---
# PrismML: 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...