[Integrations](https://spiral.ai/integrations) › [Factory Droid](https://spiral.ai/integrations/factory-droid) › [Ling 3.0 Flash VL](https://spiral.ai/integrations/factory-droid/ling-3-0-flash-vl)

# How to Use Ling 3.0 Flash VL in Factory Droid

Ling 3.0 Flash VL suits document reading, chart analysis, and tasks guided by screenshots. It adds native visual understanding to Ling 3.0 Flash and returns text from text, image, or video input.

|  |  |
| --- | --- |
| Model | Ling 3.0 Flash VL |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.000962 per run at the default inputs. $0.078 per 1M input tokens, $0.234 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Ling 3.0 Flash VL in Factory Droid

- Use Ling 3.0 Flash VL to compare the reference screenshot with our current UI screenshot. Identify three layout differences, locate the relevant components in this project, and propose the smallest fixes.
- Use Ling 3.0 Flash VL to inspect the failing browser-test screenshot. Quote any visible error message, inspect the related project code, and explain which findings are confirmed by the image.
- Use Ling 3.0 Flash VL to read the sample form images in this project. Draft an extraction schema that preserves checkbox states and marks unreadable values, then add a representative fixture.
- Use Ling 3.0 Flash VL to examine the dashboard screenshot and our chart components. Check whether displayed axis labels and units match the code, and report discrepancies with file references.
- Use Ling 3.0 Flash VL to inspect the settings-screen screenshot. Draft a browser-test sequence using visible control labels, then check those labels against the project's existing tests.

## How to set up Ling 3.0 Flash VL in Factory Droid

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

### Use Spiral's MCP server

1. **Add the server**: Droid saves it to your user config, `~/.factory/mcp.json`, so it works in every project.

```bash
droid mcp add spiral https://api.spiral.ai/mcp --type http
```

2. **Sign in**: Start `droid`, type `/mcp`, select spiral and authenticate it. Sign in to Spiral in the browser and click Authorize. No API key.

3. **Check it**: Spiral shows as connected. needs authentication means the sign-in did not finish: run `/mcp` again.

```bash
droid mcp list
```

### Use the Spiral skill

1. **Send Factory Droid 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 Ling 3.0 Flash VL**: 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": "inclusionai/ling-3.0-flash-vl",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Factory Droid finds Ling 3.0 Flash VL on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/factory-droid

## Ling 3.0 Flash VL tools and prices

- [inclusionAI: Ling 3.0 Flash VL](https://spiral.ai/tools/openrouter/inclusionai/ling-3.0-flash-vl) (OpenRouter): Up to $0.000962 per run at the default inputs. $0.078 per 1M input tokens, $0.234 per 1M output tokens.

## Settings you can ask for

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

## What can you do with Ling 3.0 Flash VL in Factory Droid?

Turn screenshots and document images into concrete development tasks with Ling 3.0 Flash VL. Ask the coding agent to connect visible evidence to project files and verify proposed changes.

### Transcribe scanned correspondence

Use Ling 3.0 Flash VL to draft a transcription that preserves spelling and reading order. Check names, dates, and identifiers against the scan.

### Extract document tables

Ask Ling 3.0 Flash VL to turn a table image into rows and columns while retaining headers, units, and footnotes.

### Explain chart changes

Have Ling 3.0 Flash VL compare chart images and write a brief that separates visible values from interpretation.

### Triage interface bugs

Use Ling 3.0 Flash VL to inspect a screenshot for overlapping controls, clipped labels, and confusing layouts, then draft a bug report.

### Draft interface instructions

Ask Ling 3.0 Flash VL to describe the visible controls needed to complete a task, citing their labels in order.

### Count visible objects

Give Ling 3.0 Flash VL an image and a precise counting rule, such as counting only fully visible boxes. Request a separate list of ambiguous objects.

### Summarize recorded demonstrations

Use Ling 3.0 Flash VL to draft a written account of the visible steps in a product demonstration, with unclear actions marked for review.

## Overview

- [OpenRouter](https://spiral.ai/providers/openrouter)
- [All Factory Droid guides](https://spiral.ai/integrations/factory-droid)

## Ling 3.0 Flash VL in other agents

- [Amp](https://spiral.ai/integrations/amp/ling-3-0-flash-vl)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/ling-3-0-flash-vl)
- [Claude](https://spiral.ai/integrations/claude-desktop/ling-3-0-flash-vl)
- [Claude Code](https://spiral.ai/integrations/claude-code/ling-3-0-flash-vl)
- [Cline](https://spiral.ai/integrations/cline/ling-3-0-flash-vl)
- [Codex](https://spiral.ai/integrations/codex/ling-3-0-flash-vl)
- [Cursor](https://spiral.ai/integrations/cursor/ling-3-0-flash-vl)
- [Devin](https://spiral.ai/integrations/devin/ling-3-0-flash-vl)
- [dots](https://spiral.ai/integrations/dots/ling-3-0-flash-vl)
- [Gemini](https://spiral.ai/integrations/gemini/ling-3-0-flash-vl)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/ling-3-0-flash-vl)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/ling-3-0-flash-vl)
- [goose](https://spiral.ai/integrations/goose/ling-3-0-flash-vl)
- [Grok](https://spiral.ai/integrations/grok/ling-3-0-flash-vl)
- [Grok Build](https://spiral.ai/integrations/grok-build/ling-3-0-flash-vl)
- [Hermes](https://spiral.ai/integrations/hermes/ling-3-0-flash-vl)
- [Instinct](https://spiral.ai/integrations/instinct/ling-3-0-flash-vl)
- [Manus](https://spiral.ai/integrations/manus/ling-3-0-flash-vl)
- [Muse](https://spiral.ai/integrations/muse/ling-3-0-flash-vl)
- [Muse Code](https://spiral.ai/integrations/muse-code/ling-3-0-flash-vl)
- [n8n](https://spiral.ai/integrations/n8n/ling-3-0-flash-vl)
- [Notion](https://spiral.ai/integrations/notion/ling-3-0-flash-vl)
- [OpenClaw](https://spiral.ai/integrations/openclaw/ling-3-0-flash-vl)
- [opencode](https://spiral.ai/integrations/opencode/ling-3-0-flash-vl)
- [Perplexity](https://spiral.ai/integrations/perplexity/ling-3-0-flash-vl)
- [Raycast](https://spiral.ai/integrations/raycast/ling-3-0-flash-vl)
- [VS Code](https://spiral.ai/integrations/vscode/ling-3-0-flash-vl)
- [Windsurf](https://spiral.ai/integrations/windsurf/ling-3-0-flash-vl)
- [Zapier](https://spiral.ai/integrations/zapier/ling-3-0-flash-vl)

## More Factory Droid guides

- [Ling 3.0 Flash](https://spiral.ai/integrations/factory-droid/ling-3-0-flash)
- [Ling 3.0 Tiny](https://spiral.ai/integrations/factory-droid/ling-3-0-tiny)
- [LFM3-1.7B](https://spiral.ai/integrations/factory-droid/lfm3-1-7b)
- [Ling 3.1 Flash](https://spiral.ai/integrations/factory-droid/ling-3-1-flash)
- [LFM2.5-24B-A2B](https://spiral.ai/integrations/factory-droid/lfm2-5-24b-a2b)
- [Llama 3 8B Lunaris](https://spiral.ai/integrations/factory-droid/llama-3-8b-lunaris)

## Questions about Ling 3.0 Flash VL in Factory Droid

**Can Factory Droid use Ling 3.0 Flash VL?**

Yes. Add Spiral to Factory Droid as an MCP server, and Factory Droid can find and run Ling 3.0 Flash VL like any built-in tool.

**How much does Ling 3.0 Flash VL cost in Factory Droid?**

Up to $0.000962 per run at the default inputs. $0.078 per 1M input tokens, $0.234 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Factory Droid plan is separate.

**Do I need an OpenRouter account to use Ling 3.0 Flash VL in Factory Droid?**

No. Spiral runs Ling 3.0 Flash VL on its own OpenRouter account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Ling 3.0 Flash VL in Factory Droid?**

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

**Does Factory Droid wait while Ling 3.0 Flash VL runs?**

Ling 3.0 Flash VL runs in the background. Factory Droid starts the run, checks on it and gives you the text, json when it's ready.

**What is Ling 3.0 Flash VL useful for?**

It handles documents, charts, object counting, and complex visual layouts. Try it for extracting a table, explaining a chart, or interpreting a software screenshot.

**Does Ling 3.0 Flash VL generate images or videos?**

It produces text from text, image, and video input. Its visual capabilities support analysis and written answers.

**Can I trust Ling 3.0 Flash VL to copy scanned text exactly?**

Check important values against the scan. OpenRouter's test recorded 254 of 264 key values correct with reasoning off, so transcription errors remain possible.

**Can Ling 3.0 Flash VL compare two images?**

Yes, multiple images can be included in one request. Label them clearly and specify what to compare, such as changed labels, chart values, or layout differences.

**How much text can Ling 3.0 Flash VL handle?**

OpenRouter currently lists a 131,072-token context window and a maximum of 32,768 completion tokens. Request shorter answers when the task only needs a few extracted values.

