[Integrations](https://spiral.ai/integrations) › [Google Antigravity](https://spiral.ai/integrations/antigravity) › [Pixtral Large 2502](https://spiral.ai/integrations/antigravity/pixtral-large-2502)

# How to Use Pixtral Large 2502 in Google Antigravity

Pixtral Large 2502 reads documents, charts, and photos and answers questions about them in text. It follows Pixtral Large 2411 and suits tasks that combine visual evidence with written context.

|  |  |
| --- | --- |
| Model | Pixtral Large 2502 |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.03204 per run at the default inputs. $2.60 per 1M input tokens, $7.80 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Pixtral Large 2502 in Google Antigravity

- Use Pixtral Large 2502 to compare the reference screenshot with the current application screenshot in this project. List visible differences in layout, labels, and controls, then identify the source files most likely responsible.
- Use Pixtral Large 2502 to read the architecture diagram and README in this repository. Explain each depicted component and list discrepancies between the diagram and the written description.
- Use Pixtral Large 2502 to inspect the chart images in the experiment reports. Identify unusual changes, quote the legend labels, and distinguish printed values from visual estimates.
- Use Pixtral Large 2502 to explain the invoice extraction code and compare its expected output with the sample invoice images. List missing cases and propose focused checks for unreadable or absent values.
- Use Pixtral Large 2502 to review the French and English help text in this project. Report differences in meaning and propose English revisions that preserve the documented behavior.

## How to set up Pixtral Large 2502 in Google Antigravity

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

### Use Spiral's MCP server

1. **Open the MCP config**: * Antigravity App: click Customizations, then Installed. In the MCP Servers section, click Copy Config File Path and open that file. * Antigravity IDE: click … at the top of the agent side panel, select MCP Servers, then Manage MCP Servers › View raw config. * Antigravity CLI: edit `~/.gemini/config/mcp_config.json`.

2. **Add the server**: Add Spiral under `mcpServers`:

```json
{
  "mcpServers": {
    "spiral": {
      "serverUrl": "https://api.spiral.ai/mcp"
    }
  }
}
```

3. **Sign in**: In the Antigravity App, open Customizations › Installed. Under MCP Servers, click Authenticate next to Spiral. Sign in to Spiral, pick the organization and click Authorize. Copy the authorization code the browser shows, paste it into Enter authorization code and click Submit. No API key.

### Use the Spiral skill

1. **Send Google Antigravity 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 Pixtral Large 2502**: 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": "mistralai/pixtral-large-2502",
    "input": {
      "messages": [
        {
          "role": "user",
          "content": "Hello!"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Google Antigravity finds Pixtral Large 2502 on Spiral, shows the price and runs it.

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

## Pixtral Large 2502 tools and prices

- [Mistral: Pixtral Large 2502](https://spiral.ai/tools/openrouter/mistralai/pixtral-large-2502) (OpenRouter): Up to $0.03204 per run at the default inputs. $2.60 per 1M input tokens, $7.80 per 1M output tokens.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| stop |  | — |  |
| top_k | 0 to … | — |  |
| top_p | 0 to 1 | — |  |
| max_tokens | 1 to 102400 | — | The most output tokens to generate. On a reasoning model the reasoning tokens count toward it: a cap too low for the effort can end with no answer (finish_reason length). |
| temperature | 0 to 2 | — |  |

## What can you do with Pixtral Large 2502 in Google Antigravity?

Turn project screenshots, diagrams, and written requirements into concrete review notes with Pixtral Large. Ask for evidence tied to the relevant source files before making changes.

### Receipt transcription

Use Pixtral Large 2502 to turn receipt images into a table of merchants, dates, items, and totals, with unreadable entries marked.

### Chart explanations

Ask Pixtral Large 2502 to identify trends and unusual changes in a chart, citing the visible labels and separating approximate readings from printed values.

### Screenshot questions

Use Pixtral Large 2502 to answer questions about visible interface text, error messages, and controls in a screenshot.

### Image comparison

Have Pixtral Large 2502 compare two product photos or document images and list visible differences for review.

### Photo captions

Ask Pixtral Large 2502 for captions grounded in visible subjects and actions, without adding identities or events absent from the image.

### Multilingual document review

Use Pixtral Large 2502 to explain French, German, or Spanish document content in English while preserving names and quoted source text.

### Questions across report pages

Give Pixtral Large 2502 relevant report text and page images, then request an answer tied to specific passages and figures.

### Code explanations

Use Pixtral Large 2502 to explain supplied code and connect its behavior to written requirements or an accompanying diagram.

## Overview

- [OpenRouter](https://spiral.ai/providers/openrouter)
- [All Google Antigravity guides](https://spiral.ai/integrations/antigravity)

## Pixtral Large 2502 in other agents

- [Amp](https://spiral.ai/integrations/amp/pixtral-large-2502)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/pixtral-large-2502)
- [Claude](https://spiral.ai/integrations/claude-desktop/pixtral-large-2502)
- [Claude Code](https://spiral.ai/integrations/claude-code/pixtral-large-2502)
- [Cline](https://spiral.ai/integrations/cline/pixtral-large-2502)
- [Codex](https://spiral.ai/integrations/codex/pixtral-large-2502)
- [Cursor](https://spiral.ai/integrations/cursor/pixtral-large-2502)
- [Devin](https://spiral.ai/integrations/devin/pixtral-large-2502)
- [dots](https://spiral.ai/integrations/dots/pixtral-large-2502)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/pixtral-large-2502)
- [Gemini](https://spiral.ai/integrations/gemini/pixtral-large-2502)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/pixtral-large-2502)
- [goose](https://spiral.ai/integrations/goose/pixtral-large-2502)
- [Grok](https://spiral.ai/integrations/grok/pixtral-large-2502)
- [Grok Build](https://spiral.ai/integrations/grok-build/pixtral-large-2502)
- [Hermes](https://spiral.ai/integrations/hermes/pixtral-large-2502)
- [Instinct](https://spiral.ai/integrations/instinct/pixtral-large-2502)
- [Manus](https://spiral.ai/integrations/manus/pixtral-large-2502)
- [Muse](https://spiral.ai/integrations/muse/pixtral-large-2502)
- [Muse Code](https://spiral.ai/integrations/muse-code/pixtral-large-2502)
- [n8n](https://spiral.ai/integrations/n8n/pixtral-large-2502)
- [Notion](https://spiral.ai/integrations/notion/pixtral-large-2502)
- [OpenClaw](https://spiral.ai/integrations/openclaw/pixtral-large-2502)
- [opencode](https://spiral.ai/integrations/opencode/pixtral-large-2502)
- [Perplexity](https://spiral.ai/integrations/perplexity/pixtral-large-2502)
- [Raycast](https://spiral.ai/integrations/raycast/pixtral-large-2502)
- [VS Code](https://spiral.ai/integrations/vscode/pixtral-large-2502)
- [Windsurf](https://spiral.ai/integrations/windsurf/pixtral-large-2502)
- [Zapier](https://spiral.ai/integrations/zapier/pixtral-large-2502)

## More Google Antigravity guides

- [Phi-4](https://spiral.ai/integrations/antigravity/phi-4)
- [Qwen Plus](https://spiral.ai/integrations/antigravity/qwen-plus)
- [Perceptron Mk1.5](https://spiral.ai/integrations/antigravity/perceptron-mk1-5)
- [Qwen2.5 72B Instruct](https://spiral.ai/integrations/antigravity/qwen2-5-72b-instruct)
- [Perceptron Mk1](https://spiral.ai/integrations/antigravity/perceptron-mk1)
- [Qwen2.5 Coder 32B Instruct](https://spiral.ai/integrations/antigravity/qwen2-5-coder-32b-instruct)

## Questions about Pixtral Large 2502 in Google Antigravity

**Can Google Antigravity use Pixtral Large 2502?**

Yes. Add Spiral to Google Antigravity as an MCP server, and Google Antigravity can find and run Pixtral Large 2502 like any built-in tool.

**How much does Pixtral Large 2502 cost in Google Antigravity?**

Up to $0.03204 per run at the default inputs. $2.60 per 1M input tokens, $7.80 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Google Antigravity plan is separate.

**Do I need an OpenRouter account to use Pixtral Large 2502 in Google Antigravity?**

No. Spiral runs Pixtral Large 2502 on its own OpenRouter account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Pixtral Large 2502 in Google Antigravity?**

About two minutes: open the MCP config, then add the server, then sign in.

**Does Google Antigravity wait while Pixtral Large 2502 runs?**

Pixtral Large 2502 runs in the background. Google Antigravity starts the run, checks on it and gives you the text, json when it's ready.

**What is Pixtral Large good at?**

Pixtral Large suits questions that combine visual evidence with text: interpreting charts, reading documents, and explaining photos. Give it a specific question rather than asking for a general description.

**Can Pixtral Large generate images?**

Pixtral Large 2502 produces text. Use it to describe an image, compare images, or draft captions.

**Which languages does Pixtral Large support?**

Documented languages include English, French, German, Spanish, Italian, Chinese, Japanese, Korean, Portuguese, Dutch, and Polish. Specify the language you want in the answer.

**How much context can Pixtral Large handle?**

Pixtral Large 2502 has a 128K-token context window. Leave room for the response when assembling a long document question.

**Where can Pixtral Large lose visual detail?**

Mistral warns that internal resizing can lose detail in very small images. For small labels or dense tables, try a readable crop and ask the model to mark unreadable text.

