[Integrations](https://spiral.ai/integrations) › [Muse Code](https://spiral.ai/integrations/muse-code) › [Qwen3-VL 8B Instruct](https://spiral.ai/integrations/muse-code/qwen3-vl-8b-instruct)

# How to Use Qwen3-VL 8B Instruct in Muse Code

Use Qwen3-VL 8B Instruct to read document images, extract text, and turn screenshots into HTML. Unlike Qwen3-VL 8B Thinking, it uses non-thinking instruction tuning; Qwen reports stronger complex visual reasoning from its Thinking models.

|  |  |
| --- | --- |
| Model | Qwen3-VL 8B Instruct |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.004006 per run at the default inputs. $0.325 per 1M input tokens, $0.975 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Qwen3-VL 8B Instruct in Muse Code

- Use Qwen3-VL 8B Instruct to turn the reference screenshot in this project into an HTML and CSS draft. Follow the project's existing component conventions and list details the screenshot leaves unclear.
- Use Qwen3-VL 8B Instruct to read the chart image in this project and draft a Python plotting script. Preserve visible labels and distinguish printed data from estimated points.
- Use Qwen3-VL 8B Instruct to compare the project's reference and current screenshots, labeled A and B. List visible differences in text, controls, and spacing before proposing code changes.
- Use Qwen3-VL 8B Instruct to transcribe the document image in this project into Markdown. Preserve heading order and mark unreadable text instead of completing it.
- Use Qwen3-VL 8B Instruct to inspect the supplied application screenshot and draft a test checklist for its visible controls. Separate what the screen shows from behavior that needs a running application to verify.

## How to set up Qwen3-VL 8B Instruct in Muse Code

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

### Use Spiral's MCP server

1. **Install Muse Code**

```bash
curl -fsSL https://dev.meta.ai/install.sh | bash
```

2. **Ask Muse to add the server**: Start `muse` and ask:

```text
Add the Spiral MCP server at https://api.spiral.ai/mcp
```

3. **Sign in**: In your terminal, run:

```bash
muse mcp login spiral
```

### Use the Spiral skill

1. **Send Muse Code 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 Qwen3-VL 8B Instruct**: 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": "qwen/qwen3-vl-8b-instruct",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Muse Code finds Qwen3-VL 8B Instruct on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/muse-code

## Qwen3-VL 8B Instruct tools and prices

- [Qwen: Qwen3 VL 8B Instruct](https://spiral.ai/tools/openrouter/qwen/qwen3-vl-8b-instruct) (OpenRouter): Up to $0.004006 per run at the default inputs. $0.325 per 1M input tokens, $0.975 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 Qwen3-VL 8B Instruct in Muse Code?

Turn screenshots and chart images into code you can revise in your project. Qwen3-VL 8B Instruct can draft HTML, plotting scripts, and explanations of visible interface controls.

### Multilingual text transcription

Transcribe photographed signs, labels, or document passages. Ask for the original wording and mark any unreadable text explicitly.

### Document field extraction

Extract named fields from document images into a consistent record. Specify what to return when a requested value is missing.

### Document layout reconstruction

Turn a page image into an outline that preserves headings, text blocks, and their positions on the page.

### Screenshot-to-HTML drafts

Convert a screenshot or hand-drawn interface sketch into an HTML draft for a developer to inspect and revise.

### Chart reproduction scripts

Generate plotting code from a chart image. Request visible labels and distinguish printed values from estimates.

### Interface walkthroughs

Read a screen's visible controls and draft instructions for a stated task. Ask the model to stop where the next screen would be needed.

### Screenshot comparison reports

Label screenshots A and B, then request a comparison of their visible text, controls, and layout.

## Overview

- [OpenRouter](https://spiral.ai/providers/openrouter)
- [All Muse Code guides](https://spiral.ai/integrations/muse-code)

## Qwen3-VL 8B Instruct in other agents

- [Amp](https://spiral.ai/integrations/amp/qwen3-vl-8b-instruct)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/qwen3-vl-8b-instruct)
- [Claude](https://spiral.ai/integrations/claude-desktop/qwen3-vl-8b-instruct)
- [Claude Code](https://spiral.ai/integrations/claude-code/qwen3-vl-8b-instruct)
- [Cline](https://spiral.ai/integrations/cline/qwen3-vl-8b-instruct)
- [Codex](https://spiral.ai/integrations/codex/qwen3-vl-8b-instruct)
- [Cursor](https://spiral.ai/integrations/cursor/qwen3-vl-8b-instruct)
- [Devin](https://spiral.ai/integrations/devin/qwen3-vl-8b-instruct)
- [dots](https://spiral.ai/integrations/dots/qwen3-vl-8b-instruct)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/qwen3-vl-8b-instruct)
- [Gemini](https://spiral.ai/integrations/gemini/qwen3-vl-8b-instruct)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/qwen3-vl-8b-instruct)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/qwen3-vl-8b-instruct)
- [goose](https://spiral.ai/integrations/goose/qwen3-vl-8b-instruct)
- [Grok](https://spiral.ai/integrations/grok/qwen3-vl-8b-instruct)
- [Grok Build](https://spiral.ai/integrations/grok-build/qwen3-vl-8b-instruct)
- [Hermes](https://spiral.ai/integrations/hermes/qwen3-vl-8b-instruct)
- [Instinct](https://spiral.ai/integrations/instinct/qwen3-vl-8b-instruct)
- [Manus](https://spiral.ai/integrations/manus/qwen3-vl-8b-instruct)
- [Muse](https://spiral.ai/integrations/muse/qwen3-vl-8b-instruct)
- [n8n](https://spiral.ai/integrations/n8n/qwen3-vl-8b-instruct)
- [Notion](https://spiral.ai/integrations/notion/qwen3-vl-8b-instruct)
- [OpenClaw](https://spiral.ai/integrations/openclaw/qwen3-vl-8b-instruct)
- [opencode](https://spiral.ai/integrations/opencode/qwen3-vl-8b-instruct)
- [Perplexity](https://spiral.ai/integrations/perplexity/qwen3-vl-8b-instruct)
- [Raycast](https://spiral.ai/integrations/raycast/qwen3-vl-8b-instruct)
- [VS Code](https://spiral.ai/integrations/vscode/qwen3-vl-8b-instruct)
- [Windsurf](https://spiral.ai/integrations/windsurf/qwen3-vl-8b-instruct)
- [Zapier](https://spiral.ai/integrations/zapier/qwen3-vl-8b-instruct)

## More Muse Code guides

- [Qwen3 VL 30B A3B Instruct](https://spiral.ai/integrations/muse-code/qwen3-vl-30b-a3b-instruct)
- [Qwen3.5 397B A17B](https://spiral.ai/integrations/muse-code/qwen3-5-397b-a17b)
- [Qwen3 VL 235B A22B Instruct](https://spiral.ai/integrations/muse-code/qwen3-vl-235b-a22b-instruct)
- [Qwen3.5 Plus](https://spiral.ai/integrations/muse-code/qwen3-5-plus)
- [Qwen3 Next 80B A3B Instruct](https://spiral.ai/integrations/muse-code/qwen3-next-80b-a3b-instruct)
- [Qwen3.5 Plus 2026-04-20](https://spiral.ai/integrations/muse-code/qwen3-5-plus-2026-04-20)

## Questions about Qwen3-VL 8B Instruct in Muse Code

**Can Muse Code use Qwen3-VL 8B Instruct?**

Yes. Add Spiral to Muse Code as an MCP server, and Muse Code can find and run Qwen3-VL 8B Instruct like any built-in tool.

**How much does Qwen3-VL 8B Instruct cost in Muse Code?**

Up to $0.004006 per run at the default inputs. $0.325 per 1M input tokens, $0.975 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Muse Code plan is separate.

**Do I need an OpenRouter account to use Qwen3-VL 8B Instruct in Muse Code?**

No. Spiral runs Qwen3-VL 8B Instruct on its own OpenRouter account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Qwen3-VL 8B Instruct in Muse Code?**

About two minutes: install Muse Code, then ask Muse to add the server, then sign in.

**Does Muse Code wait while Qwen3-VL 8B Instruct runs?**

Qwen3-VL 8B Instruct runs in the background. Muse Code starts the run, checks on it and gives you the text, json when it's ready.

**What is Qwen3-VL 8B Instruct useful for?**

Useful tasks include reading document images, extracting text, interpreting interface screenshots, and drafting HTML from a screenshot. Qwen also demonstrates generating plotting code from chart images.

**How many languages does its OCR support?**

Qwen documents OCR support for 32 languages. That count describes text recognition, not a guarantee of equal accuracy across languages.

**Can it turn a screenshot into code?**

Yes. Qwen demonstrates converting screenshots and sketches into HTML, plus generating plotting code from chart images.

**How much context does it have, and how long can its answers be?**

The published model configuration specifies 262,144 context tokens. Qwen uses up to 32,768 output tokens in its Instruct text evaluations, but that setting does not establish a universal answer-length limit.

**Where is Qwen3-VL 8B Instruct weaker than Thinking models?**

Complex visual reasoning is a relative weakness: Qwen reports stronger multimodal reasoning from its Thinking models. For difficult problems with several reasoning steps, treat an Instruct answer as a draft to verify.

