Ternary Bonsai 2 9B is PrismML’s compact, open-weight language model for conversational AI, reasoning, and tool use. Built on Qwen3.5, it uses ternary weights packed into 2 bits for efficient inference. It accepts text and image input. Its 262K-token context window supports long conversations and document-heavy workflows, with up to 64K output tokens per response.
$0.04
$0.02
$0.065
Throughput is how fast the model writes (tokens per second, higher is better). Time to first token is how long before the answer starts (lower is better).
Share of requests served without a server error. Client errors (4xx) do not count.
Works with the OpenAI SDK pointed at our base URL. Every parameter is in the API reference.
Tokens processed per day on NagaAI.
Ternary Bonsai 2 9B is PrismML’s compact, open-weight language model for conversational AI, reasoning, and tool use. Built on Qwen3.5, it uses ternary weights packed into 2 bits for efficient inference. It accepts text and image input.
On NagaAI, Ternary Bonsai 2 9B costs $0.04 per 1M input tokens and $0.065 per 1M output tokens. You pay only for what you use, with no subscription.
Ternary Bonsai 2 9B reads up to 262.1k tokens in one request, counting the prompt and the answer together.
Yes. Ternary Bonsai 2 9B can call your functions as tools and can answer in a JSON schema you give it.
It takes text and images as input and returns text.
Create an API key, copy the model ID from the top of this page and send requests with the OpenAI or Anthropic SDK pointed at NagaAI. The API section on this page has a ready example.
Not enough throughput data
Not enough TTFT data
Not enough activity data to display a chart
