[Integrations](https://spiral.ai/integrations) › [Amp](https://spiral.ai/integrations/amp) › [Ternary Bonsai 2 9B](https://spiral.ai/integrations/amp/ternary-bonsai-2-9b)

# How to Use Ternary Bonsai 2 9B in Amp

Ternary Bonsai 2 9B suits focused coding tasks, everyday math, and tool-assisted document work. Its 262,144-token context accommodates substantial reference material, while thinking mode needs enough output space to finish.

|  |  |
| --- | --- |
| Model | Ternary Bonsai 2 9B |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.000697 per run at the default inputs. $0.104 per 1M input tokens, $0.169 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Ternary Bonsai 2 9B in Amp

- Use Ternary Bonsai 2 9B to inspect the CSV parser and its tests in this project. Find why quoted commas are split incorrectly, make the smallest fix, and run the relevant tests.
- Use Ternary Bonsai 2 9B to draft tests for the project's pagination helper. Cover an empty result, the first page, the last page, and an out-of-range page; derive expectations from the documented contract.
- Use Ternary Bonsai 2 9B to compare the retry implementation with the project's README. List discrepancies with file references, then propose one focused correction.
- Use Ternary Bonsai 2 9B to investigate the failing unit test. Keep thinking enabled, use temperature 1.0, top_p 0.95, and top_k 20, and allow at least 8,192 output tokens before requesting a minimal patch.
- Use Ternary Bonsai 2 9B to rewrite the project's error-handling documentation in under 250 words. Preserve every exception name and include one example drawn from the implementation.

## How to set up Ternary Bonsai 2 9B in Amp

Pick the way that suits you. Each takes about two minutes.

### Use Spiral's MCP server

1. **Add the server**: No header and no client ID or secret: Amp registers itself with Spiral.

```bash
amp mcp add spiral https://api.spiral.ai/mcp
```

2. **Sign in**: Start `amp`. Amp opens Spiral's sign-in page in your browser: sign in, pick the organization and click Authorize. No API key.

3. **Check it**

```bash
amp mcp doctor
```

### Use the Spiral skill

1. **Send Amp this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **Sign in or paste a key**: Approve Spiral in the browser, or give the agent a key from the console.

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run Ternary Bonsai 2 9B**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "openrouter",
    "model": "prism-ml/ternary-bonsai-2-9b",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Amp finds Ternary Bonsai 2 9B on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/amp

## Ternary Bonsai 2 9B tools and prices

- [PrismML: Ternary Bonsai 2 9B](https://spiral.ai/tools/openrouter/prism-ml/ternary-bonsai-2-9b) (OpenRouter): Up to $0.000697 per run at the default inputs. $0.104 per 1M input tokens, $0.169 per 1M output tokens.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| seed |  | — |  |
| stop |  | — |  |
| tools |  | — |  |
| top_k | 0 to … | — |  |
| top_p | 0 to 1 | — |  |
| logprobs |  | — |  |

## What can you do with Ternary Bonsai 2 9B in Amp?

Ternary Bonsai 2 9B can help turn project code and requirements into focused patches and test drafts. Give it a bounded change, relevant files, and an observable success condition.

### Small code fixes

Ask Ternary Bonsai 2 9B to fix one function against a clear requirement. Request a minimal patch and an explanation of the behavior it changes.

### Test case drafts

Give Ternary Bonsai 2 9B a function and its expected behavior. Ask for tests covering normal inputs, boundary values, and failures.

### Everyday math checks

Use Ternary Bonsai 2 9B to work through quantities, percentages, and unit conversions. Require the formula, substituted values, and final units.

### Questions about reference text

Supply a policy or product specification to Ternary Bonsai 2 9B. Ask for an answer with supporting passages and a clear statement when the text leaves something unresolved.

### Document extraction

Ask Ternary Bonsai 2 9B to extract dates, organizations, obligations, and action items from document text. Require missing details to remain explicitly unknown.

### Constrained rewrites

Give Ternary Bonsai 2 9B a draft, a target audience, and a firm length limit. Ask it to preserve names, numbers, and commitments while removing repetition.

### Support request triage

Use Ternary Bonsai 2 9B to sort request text into a small set of defined categories. Ask for the triggering passage and send ambiguous requests for review.

### Tool-assisted document checks

Have Ternary Bonsai 2 9B extract a reference number from text, request the corresponding record through an available tool, and summarize any discrepancy.

## Overview

- [OpenRouter](https://spiral.ai/providers/openrouter)
- [All Amp guides](https://spiral.ai/integrations/amp)

## Ternary Bonsai 2 9B in other agents

- [ChatGPT](https://spiral.ai/integrations/chatgpt/ternary-bonsai-2-9b)
- [Claude](https://spiral.ai/integrations/claude-desktop/ternary-bonsai-2-9b)
- [Claude Code](https://spiral.ai/integrations/claude-code/ternary-bonsai-2-9b)
- [Cline](https://spiral.ai/integrations/cline/ternary-bonsai-2-9b)
- [Codex](https://spiral.ai/integrations/codex/ternary-bonsai-2-9b)
- [Cursor](https://spiral.ai/integrations/cursor/ternary-bonsai-2-9b)
- [Devin](https://spiral.ai/integrations/devin/ternary-bonsai-2-9b)
- [dots](https://spiral.ai/integrations/dots/ternary-bonsai-2-9b)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/ternary-bonsai-2-9b)
- [Gemini](https://spiral.ai/integrations/gemini/ternary-bonsai-2-9b)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/ternary-bonsai-2-9b)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/ternary-bonsai-2-9b)
- [goose](https://spiral.ai/integrations/goose/ternary-bonsai-2-9b)
- [Grok](https://spiral.ai/integrations/grok/ternary-bonsai-2-9b)
- [Grok Build](https://spiral.ai/integrations/grok-build/ternary-bonsai-2-9b)
- [Hermes](https://spiral.ai/integrations/hermes/ternary-bonsai-2-9b)
- [Instinct](https://spiral.ai/integrations/instinct/ternary-bonsai-2-9b)
- [Manus](https://spiral.ai/integrations/manus/ternary-bonsai-2-9b)
- [Muse](https://spiral.ai/integrations/muse/ternary-bonsai-2-9b)
- [Muse Code](https://spiral.ai/integrations/muse-code/ternary-bonsai-2-9b)
- [n8n](https://spiral.ai/integrations/n8n/ternary-bonsai-2-9b)
- [Notion](https://spiral.ai/integrations/notion/ternary-bonsai-2-9b)
- [OpenClaw](https://spiral.ai/integrations/openclaw/ternary-bonsai-2-9b)
- [opencode](https://spiral.ai/integrations/opencode/ternary-bonsai-2-9b)
- [Perplexity](https://spiral.ai/integrations/perplexity/ternary-bonsai-2-9b)
- [Raycast](https://spiral.ai/integrations/raycast/ternary-bonsai-2-9b)
- [VS Code](https://spiral.ai/integrations/vscode/ternary-bonsai-2-9b)
- [Windsurf](https://spiral.ai/integrations/windsurf/ternary-bonsai-2-9b)
- [Zapier](https://spiral.ai/integrations/zapier/ternary-bonsai-2-9b)

## More Amp guides

- [Ternary Bonsai 2 27B](https://spiral.ai/integrations/amp/ternary-bonsai-2-27b)
- [UI-TARS 1.5 7B](https://spiral.ai/integrations/amp/ui-tars-1-5-7b)
- [Step 5 Preview](https://spiral.ai/integrations/amp/step-5-preview)
- [UnslopNemo 12B](https://spiral.ai/integrations/amp/unslopnemo-12b)
- [Step 3.7 Flash](https://spiral.ai/integrations/amp/step-3-7-flash)
- [Venice Uncensored](https://spiral.ai/integrations/amp/venice-uncensored)

## Questions about Ternary Bonsai 2 9B in Amp

**Can Amp use Ternary Bonsai 2 9B?**

Yes. Add Spiral to Amp as an MCP server, and Amp can find and run Ternary Bonsai 2 9B like any built-in tool.

**How much does Ternary Bonsai 2 9B cost in Amp?**

Up to $0.000697 per run at the default inputs. $0.104 per 1M input tokens, $0.169 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Amp plan is separate.

**Do I need an OpenRouter account to use Ternary Bonsai 2 9B in Amp?**

No. Spiral runs Ternary Bonsai 2 9B on its own OpenRouter account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Ternary Bonsai 2 9B in Amp?**

About two minutes: add the server, then sign in, then check it.

**Does Amp wait while Ternary Bonsai 2 9B runs?**

Ternary Bonsai 2 9B runs in the background. Amp starts the run, checks on it and gives you the text, json when it's ready.

**What is Ternary Bonsai 2 9B good at?**

Focused coding, everyday math, and tasks with explicit instructions are sensible starting points. PrismML reports that its coding, math, and instruction-following benchmark results match or exceed the corresponding full-precision model.

**How much context can Ternary Bonsai 2 9B handle?**

The documented maximum context is 262,144 tokens. For document questions, ask it to identify the supporting passage before writing its answer.

**Why can a short question need a large output allowance?**

Thinking mode is enabled by default, and even simple prompts can consume more than 3,000 thinking tokens. PrismML recommends allowing at least 8,192 output tokens with thinking enabled.

**Which sampling settings should I start with?**

For thinking mode, start with temperature 1.0, top-p 0.95, and top-k 20. Without thinking, use temperature 0.7, top-p 0.80, and top-k 20.

**What is a documented weakness of Ternary Bonsai 2 9B?**

It can repeat or get stuck reasoning on some prompts. PrismML specifically warns against temperature 0 with thinking enabled.

