AI Agent StorePrismML · Provider and hosting offer verified
Ternary Bonsai 2 9B
Use a compact text-and-image model through OpenRouter, or download it for a compatible local setup. Its compressed files reduce storage needs; test quality and total memory on your own task.
Catalog added 2026-10-16 UTC · Checked 2026-11-03. Catalog addition and provider launch are separate dates.
We still need comparable independent tests. You can compare the confirmed features and limits below, but we cannot yet give this version an Agent Fit or Practical Value rating. Test it on your own tasks before switching; results for another version or thinking setting may not apply.
What you can use today
Token prices measure the small pieces of text the model reads or produces. Capacity is the amount the service accepts in a request; check the limits below for any fallback to another model. A dash means the rate is unverified.
Exact catalog ID: prism-ml/ternary-bonsai-2-9b. Derived from Qwen3.5-9B, with ternary weights. This model has no independent result here and does not inherit Qwen’s scores.
Apache 2.0 weights are downloadable. Use PrismML’s llama.cpp fork for its supported compressed formats; images require the separate vision file. File size is not total runtime memory.
OpenRouter lists 262,144 context tokens and up to 65,536 output tokens. The download also documents 262,144 context; its separate output maximum is not established.
Rates describe the catalog offer observed on 2026-11-03. Allow for retries and any tool fees. Models that think before answering may also charge for that work.
Provider release and specifications ↗
Inspect the exact catalog model ↗
Test it before switching
Give each model the same examples, tools and definition of success. Record correct outcomes, response time, retries and total cost. A cheaper token price does not guarantee a cheaper completed task.
Compare caching and long-context costs ↗