[Integrations](https://spiral.ai/integrations) › [Gemini CLI](https://spiral.ai/integrations/gemini-cli) › [Amazon Nova 2 Lite](https://spiral.ai/integrations/gemini-cli/amazon-nova-2-lite)

# How to Use Amazon Nova 2 Lite in Gemini CLI

Amazon Nova 2 Lite handles document questions, code review, and repeated text tasks. Use it without extended thinking for routine work, or choose medium reasoning effort for debugging and plans with several constraints.

|  |  |
| --- | --- |
| Model | Amazon Nova 2 Lite |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.013327 per run at the default inputs. $0.39 per 1M input tokens, $3.25 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Amazon Nova 2 Lite in Gemini CLI

- Use Amazon Nova 2 Lite with low reasoning effort to review the current diff for logic errors and missing edge cases. Cite file paths and lines, explain a concrete failure scenario for each finding, and omit style-only suggestions.
- Use Amazon Nova 2 Lite with medium reasoning effort to investigate the failing test using the relevant source and logs. Compare the likely causes, propose the smallest fix, and run the test that checks it.
- Use Amazon Nova 2 Lite to answer questions about this repository's authentication flow. Gather the relevant code first, label each file, then ask for the request path, permission checks, and any unresolved questions with file references.
- Use Amazon Nova 2 Lite with medium reasoning effort to compare two ways to remove the duplicated validation logic in this project. Evaluate behavior changes, dependencies, and testing needs, then recommend an implementation plan.
- Use Amazon Nova 2 Lite without extended thinking to draft release notes from the current diff and commit messages. Separate user-visible changes from internal maintenance, and include only behavior supported by those sources.

## How to set up Amazon Nova 2 Lite in Gemini CLI

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

### Use Spiral's MCP server

1. **Add the server**: Or add it to `~/.gemini/settings.json` yourself:

```bash
gemini mcp add --scope user --transport http spiral https://api.spiral.ai/mcp
```

2. **Sign in**: Start `gemini` and run:

```text
/mcp auth spiral
```

3. **Check it**: In Gemini CLI, run `/mcp`. From your shell:

```bash
gemini mcp list
```

### Use the Spiral skill

1. **Send Gemini CLI 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 Amazon Nova 2 Lite**: 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": "amazon/nova-2-lite-v1",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Gemini CLI finds Amazon Nova 2 Lite on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/gemini-cli

## Amazon Nova 2 Lite tools and prices

- [Amazon: Nova 2 Lite](https://spiral.ai/tools/openrouter/amazon/nova-2-lite-v1) (OpenRouter): Up to $0.013327 per run at the default inputs. $0.39 per 1M input tokens, $3.25 per 1M output tokens.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| stop |  | — |  |
| tools |  | — |  |
| top_k | 0 to … | — |  |
| top_p | 0 to 1 | — |  |
| reasoning |  | — | Reasoning effort; reasoning tokens count toward max_tokens. |
| max_tokens | 1 to 65535 | — | The most output tokens to generate. |

## What can you do with Amazon Nova 2 Lite in Gemini CLI?

Review changes, trace bugs, and compare implementation plans against the project's code with Amazon Nova 2 Lite. Use low reasoning effort for reviews and medium effort when the task spans several components.

### Questions over long reference material

Use Amazon Nova 2 Lite to answer questions across manuals or reports. Ask for a supporting passage and document identifier for each answer.

### Code review

Ask Amazon Nova 2 Lite to inspect a proposed change for logic errors and missing edge cases. Use low reasoning effort and request concrete examples of possible failures.

### Debugging across components

Give Amazon Nova 2 Lite relevant code, logs, and expected behavior. Use medium reasoning effort to compare possible causes and propose a focused fix.

### Meeting recaps

Turn a supplied meeting transcript into decisions, unresolved questions, and assigned actions with Amazon Nova 2 Lite. Ask it to mark owners and deadlines as unknown when absent.

### Support ticket classification

Use Amazon Nova 2 Lite to sort incoming messages into defined categories and summarize each request. Include an uncertain category for messages that do not clearly fit.

### Comparing proposals

Ask Amazon Nova 2 Lite to compare written proposals against explicit requirements. Use medium reasoning effort and require evidence for each stated advantage or gap.

### Structured document summaries

Have Amazon Nova 2 Lite turn supplied text into JSON containing a summary, key dates, and open questions. Define how it should represent missing information.

### Multilingual feedback summaries

Use Amazon Nova 2 Lite to summarize English, Spanish, French, or Japanese customer feedback in English. Keep an original-language excerpt beside each theme for review.

## Overview

- [OpenRouter](https://spiral.ai/providers/openrouter)
- [All Gemini CLI guides](https://spiral.ai/integrations/gemini-cli)

## Amazon Nova 2 Lite in other agents

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

## More Gemini CLI guides

- [Aion-RP 1.0 (8B)](https://spiral.ai/integrations/gemini-cli/aion-rp-1-0-8b)
- [Amazon Nova Lite 1.0](https://spiral.ai/integrations/gemini-cli/amazon-nova-lite-1-0)
- [Aion 3.5 Mini](https://spiral.ai/integrations/gemini-cli/aion-3-5-mini)
- [Amazon Nova Micro](https://spiral.ai/integrations/gemini-cli/amazon-nova-micro)
- [Aion 3.5](https://spiral.ai/integrations/gemini-cli/aion-3-5)
- [Amazon Nova Premier 1.0](https://spiral.ai/integrations/gemini-cli/amazon-nova-premier-1-0)

## Questions about Amazon Nova 2 Lite in Gemini CLI

**Can Gemini CLI use Amazon Nova 2 Lite?**

Yes. Add Spiral to Gemini CLI as an MCP server, and Gemini CLI can find and run Amazon Nova 2 Lite like any built-in tool.

**How much does Amazon Nova 2 Lite cost in Gemini CLI?**

Up to $0.013327 per run at the default inputs. $0.39 per 1M input tokens, $3.25 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Gemini CLI plan is separate.

**Do I need an OpenRouter account to use Amazon Nova 2 Lite in Gemini CLI?**

No. Spiral runs Amazon Nova 2 Lite on its own OpenRouter account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Amazon Nova 2 Lite in Gemini CLI?**

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

**Does Gemini CLI wait while Amazon Nova 2 Lite runs?**

Amazon Nova 2 Lite runs in the background. Gemini CLI starts the run, checks on it and gives you the text, json when it's ready.

**What is Amazon Nova 2 Lite good at?**

Start with document summaries, text classification, and questions over supplied reference material. Amazon also recommends its reasoning modes for code review, debugging, and planning with multiple constraints.

**How much context and output can Amazon Nova 2 Lite handle?**

Amazon's overview lists up to 1 million tokens of context and 65,536 generated tokens per response. Its extended-thinking guide separately describes outputs above 65,000 tokens at high reasoning effort.

**When should I use extended thinking?**

Leave it off for routine text work. Amazon recommends low effort for code review and medium effort for debugging, development across multiple components, and planning with several constraints.

**Which languages does Amazon Nova 2 Lite support?**

It supports over 200 languages, with optimization for 15: English, German, Spanish, French, Italian, Japanese, Korean, Arabic, Simplified Chinese, Russian, Hindi, Portuguese, Dutch, Turkish, and Hebrew.

**Can Amazon Nova 2 Lite return JSON?**

Yes. State the required structure and ask for JSON only; include rules for missing values and permitted labels so the requested output is unambiguous.

