[Integrations](https://spiral.ai/integrations) › [Amp](https://spiral.ai/integrations/amp) › [Command R7B (12-2024)](https://spiral.ai/integrations/amp/command-r7b-12-2024)

# How to Use Command R7B (12-2024) in Amp

Command R7B (12-2024) handles document Q&A, summaries, and code assistance. Its 128,000-token context supports long inputs, while replies are capped at 4,000 tokens.

|  |  |
| --- | --- |
| Model | Command R7B (12-2024) |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.000783 per run at the default inputs. $0.04875 per 1M input tokens, $0.195 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Command R7B (12-2024) in Amp

- Use Command R7B (12-2024) to explain the authentication flow in this project's source files. Cite file paths and function names, distinguish observed behavior from assumptions, and keep the explanation under 600 words.
- Use Command R7B (12-2024) with low temperature to draft a PostgreSQL query from this project's schema. Find customers with no orders in the last 90 days, explain the joins, and flag any missing schema details before guessing.
- Use Command R7B (12-2024) to translate the selected JavaScript utility into Python. Preserve its documented behavior, use the existing test examples as checks, and return one function plus a short explanation.
- Use Command R7B (12-2024) to answer onboarding questions from the project's README and documentation. Select relevant excerpts of roughly 100 to 400 words, cite their file paths, and say when the documentation does not answer a question.
- Use Command R7B (12-2024) to categorize the TODO comments in this repository as bug, documentation, refactor, or uncertain. Return each comment with its path, category, and a one-sentence reason.

## How to set up Command R7B (12-2024) 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 Command R7B (12-2024)**: 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": "cohere/command-r7b-12-2024",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Amp finds Command R7B (12-2024) on Spiral, shows the price and runs it.

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

## Command R7B (12-2024) tools and prices

- [Cohere: Command R7B (12-2024)](https://spiral.ai/tools/openrouter/cohere/command-r7b-12-2024) (OpenRouter): Up to $0.000783 per run at the default inputs. $0.04875 per 1M input tokens, $0.195 per 1M output tokens.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| seed |  | — |  |
| stop |  | — |  |
| top_k | 0 to … | — |  |
| top_p | 0 to 1 | — |  |
| max_tokens | 1 to 4000 | — | The most output tokens to generate. |
| temperature | 0 to 2 | — |  |

## What can you do with Command R7B (12-2024) in Amp?

Turn project code and documentation into explanations, SQL drafts, and small code rewrites with Command R7B. Supply relevant files and expected behavior, and keep each requested change small enough for its 4,000-token output limit.

### Answer policy questions

Give Command R7B labeled policy excerpts and ask for an answer tied to those sources. Request a list of missing details when the policy does not resolve the question.

### Summarize meeting transcripts

Turn a text transcript into decisions, action items, and open questions. Ask Command R7B to preserve stated owners and deadlines without filling in missing ones.

### Extract contract details

Ask Command R7B to pull renewal dates, notice periods, and named parties from contract text. Include an exact supporting excerpt for each extracted value.

### Translate support replies

Use Command R7B to translate support text between its trained languages. Supply a glossary and require product names, dates, and reference numbers to remain unchanged.

### Categorize customer requests

Give Command R7B a fixed set of categories and short definitions, then ask it to label incoming requests. Include an uncertain category for messages that do not clearly fit.

### Draft SQL from a schema

Provide table definitions and ask Command R7B for a query with its assumptions. Specify the SQL dialect and request an explanation of joins and filters.

### Explain or translate code

Ask Command R7B to explain a function or translate a small routine into another programming language. Supply expected behavior and examples to check against the result.

### Prepare a document digest

Use Command R7B to summarize supplied document excerpts into a short briefing. Ask for source labels beside each point and separate unresolved questions from reported facts.

## Overview

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

## Command R7B (12-2024) in other agents

- [ChatGPT](https://spiral.ai/integrations/chatgpt/command-r7b-12-2024)
- [Claude](https://spiral.ai/integrations/claude-desktop/command-r7b-12-2024)
- [Claude Code](https://spiral.ai/integrations/claude-code/command-r7b-12-2024)
- [Cline](https://spiral.ai/integrations/cline/command-r7b-12-2024)
- [Codex](https://spiral.ai/integrations/codex/command-r7b-12-2024)
- [Cursor](https://spiral.ai/integrations/cursor/command-r7b-12-2024)
- [Devin](https://spiral.ai/integrations/devin/command-r7b-12-2024)
- [dots](https://spiral.ai/integrations/dots/command-r7b-12-2024)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/command-r7b-12-2024)
- [Gemini](https://spiral.ai/integrations/gemini/command-r7b-12-2024)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/command-r7b-12-2024)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/command-r7b-12-2024)
- [goose](https://spiral.ai/integrations/goose/command-r7b-12-2024)
- [Grok](https://spiral.ai/integrations/grok/command-r7b-12-2024)
- [Grok Build](https://spiral.ai/integrations/grok-build/command-r7b-12-2024)
- [Hermes](https://spiral.ai/integrations/hermes/command-r7b-12-2024)
- [Instinct](https://spiral.ai/integrations/instinct/command-r7b-12-2024)
- [Manus](https://spiral.ai/integrations/manus/command-r7b-12-2024)
- [Muse](https://spiral.ai/integrations/muse/command-r7b-12-2024)
- [Muse Code](https://spiral.ai/integrations/muse-code/command-r7b-12-2024)
- [n8n](https://spiral.ai/integrations/n8n/command-r7b-12-2024)
- [Notion](https://spiral.ai/integrations/notion/command-r7b-12-2024)
- [OpenClaw](https://spiral.ai/integrations/openclaw/command-r7b-12-2024)
- [opencode](https://spiral.ai/integrations/opencode/command-r7b-12-2024)
- [Perplexity](https://spiral.ai/integrations/perplexity/command-r7b-12-2024)
- [Raycast](https://spiral.ai/integrations/raycast/command-r7b-12-2024)
- [VS Code](https://spiral.ai/integrations/vscode/command-r7b-12-2024)
- [Windsurf](https://spiral.ai/integrations/windsurf/command-r7b-12-2024)
- [Zapier](https://spiral.ai/integrations/zapier/command-r7b-12-2024)

## More Amp guides

- [Command R Plus (08-2024)](https://spiral.ai/integrations/amp/command-r-plus-08-2024)
- [Cydonia 24B V4.1](https://spiral.ai/integrations/amp/cydonia-24b-v4-1)
- [Command R (08-2024)](https://spiral.ai/integrations/amp/command-r-08-2024)
- [DaisyRose 27B](https://spiral.ai/integrations/amp/daisyrose-27b)
- [Command A Plus](https://spiral.ai/integrations/amp/command-a-plus)
- [DeepSeek Flash Latest](https://spiral.ai/integrations/amp/deepseek-flash-latest)

## Questions about Command R7B (12-2024) in Amp

**Can Amp use Command R7B (12-2024)?**

Yes. Add Spiral to Amp as an MCP server, and Amp can find and run Command R7B (12-2024) like any built-in tool.

**How much does Command R7B (12-2024) cost in Amp?**

Up to $0.000783 per run at the default inputs. $0.04875 per 1M input tokens, $0.195 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 Command R7B (12-2024) in Amp?**

No. Spiral runs Command R7B (12-2024) on its own OpenRouter account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Command R7B (12-2024) in Amp?**

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

**Does Amp wait while Command R7B (12-2024) runs?**

Command R7B (12-2024) runs in the background. Amp starts the run, checks on it and gives you the text, json when it's ready.

**What is Command R7B good at?**

Use it to answer questions from supplied documents, summarize text, extract details, translate, and categorize content. Cohere also identifies code assistance as a suitable use.

**How much text can Command R7B handle?**

It has a 128,000-token context window and can generate up to 4,000 tokens per response. For a long report, request separate sections so each reply fits the output limit.

**Which languages does Command R7B support?**

It was trained on 23 languages, including English, French, Spanish, German, Japanese, Korean, Arabic, Chinese, and Hindi. Specify the output language and identify any names or technical terms that should stay unchanged.

**How should I prepare documents for Command R7B?**

Cohere recommends short document snippets of roughly 100 to 400 words for grounded generation. Give each excerpt a source label, ask a specific question, and request an explicit statement when the excerpts lack the answer.

**Can Command R7B help with SQL and code?**

Yes. Cohere describes SQL, code translation, snippets, explanations, and rewrites as supported uses, and recommends low temperature for code generation.

