Ternary Bonsai 2 9B
Toolsprism-ml/ternary-bonsai-2-9b262K context · Oct 2026 · 1 upstreamsupstream上游
Bonsai 2 9B is PrismML's compact open-weight vision-language model, derived from Qwen3.5-9B with ternary weights. It handles conversation, reasoning and tool use, takes image input, and has a 262K-token context window.
Needs your own keyTo use Ternary Bonsai 2 9B, add a key from OpenRouter. AnyRouter adds no fee.
- contextngữ cảnh上下文
- 262K
- inputđầu vào输入 / 1M (list)
- $0.08
- outputđầu ra输出 / 1M (list)
- $0.13
- routestuyến路由
- 1
- speed, 24htốc độ, 24h速度,24h
- —
Upstreams
1- 1OpenRouterYour key · not addedyour planTPS —TTFT ——
- Up
- No traffic
- Down
Details
- Owner
- Prism Ml
- Kind
- Vision
- Released
- Oct 2026
- Context
- 262K
- Data policy
- Provider retention
- Endpoint
/api/v1/chat/completions
Quick start
curl https://anyrouter.dev/api/v1/chat/completions \ -H "Authorization: Bearer $ANYROUTER_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "prism-ml/ternary-bonsai-2-9b", "messages": [ { "role": "user", "content": "Say hi in 3 words." } ] }'
- Low-cost chat and tool-calling agents
- Image understanding with text output
Architecture and licenseKiến trúc và giấy phép架构与许可
- TypeLoại类型
- Dense[1]
- Native contextNgữ cảnh gốc原生上下文
- 262,144[1]
- LicenseGiấy phép许可
- Apache-2.0 · Open weightsTrọng số mở开放权重[1]
StrengthsĐiểm mạnh优势[1]
- Ternary weights shrink the language model to about 2.4 GB from about 17.9 GB at FP16
- Image input through a separate vision projector
- 262K-token context from the Qwen3.5-9B hybrid-attention backbone
Known limitsGiới hạn đã biết已知局限[1]
- Thinking mode can use 3,000+ tokens on simple prompts, so allow at least 8,192 output tokens
- Greedy decoding with thinking on can get stuck reasoning on some prompts
API notesGhi chú APIAPI 说明[1]
- Card recommends temperature 1.0, top_p 0.95, top_k 20 and presence_penalty 1.0 in thinking mode
- Thinking can be turned off per request with chat_template_kwargs enable_thinking false
Links and versionsLiên kết và phiên bản链接与版本
- Model cardThẻ mô hình模型卡
- huggingface.co
- Previous versionPhiên bản trước上一版本
- prism-ml/ternary-bonsai-2-27b
SourcesNguồn来源
- [1]prism-ml/Ternary-Bonsai-2-9B-gguf model cardaccessedtruy cập访问于 2026-10-28
Upstreams and pricingUpstream và giá上游与价格
Share cards2 images


params: frequency_penalty · include_reasoning · logprobs · max_tokens · presence_penalty · reasoning · repetition_penalty · response_format · seed · stop · structured_outputs · temperature · tool_choice · tools · top_k · top_logprobs · top_p
Uptime + latency
—recent checks · all routes
No health checks recorded for this model yet.
24h
No traffic in the last 24h.