[Integrations](https://spiral.ai/integrations) › [opencode](https://spiral.ai/integrations/opencode) › [Qwen2.5 72B Instruct](https://spiral.ai/integrations/opencode/qwen2-5-72b-instruct)

# How to Use Qwen2.5 72B Instruct in opencode

Qwen2.5 72B Instruct handles multilingual writing, table analysis, JSON extraction, and code. Compared with Qwen2 72B Instruct, Qwen reports better math and coding results, making it useful for work that mixes prose with technical tasks.

|  |  |
| --- | --- |
| Model | Qwen2.5 72B Instruct |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.002147 per run at the default inputs. $0.468 per 1M input tokens, $0.52 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Qwen2.5 72B Instruct in opencode

- Use Qwen2.5 72B Instruct to review the current diff for incorrect behavior. Return at most five findings with file paths, supporting code, and a test that would expose each issue.
- Use Qwen2.5 72B Instruct to draft tests for the CSV parser in this project. Cover quoted commas, empty cells, malformed rows, and Unicode text; follow the existing test conventions and run the relevant tests.
- Use Qwen2.5 72B Instruct to compare the API specification with the request validation code. Return JSON containing the endpoint, mismatch, supporting file locations, and proposed correction.
- Use Qwen2.5 72B Instruct to translate the English help strings in this project into Spanish and Japanese. Preserve keys, interpolation tokens, product names, and Markdown formatting; report ambiguous terms.
- Use Qwen2.5 72B Instruct with temperature 0.7, top-p 0.8, and top-k 20 to draft a Python function from the selected specification. Include type hints and focused tests, then check the result against the specification.

## How to set up Qwen2.5 72B Instruct in opencode

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

### Use Spiral's MCP server

1. **Add the server**: Leave out `--global` to add Spiral to the current project only. No header and no client ID or secret: opencode registers itself with Spiral.

```bash
opencode mcp add spiral --global --url https://api.spiral.ai/mcp
```

2. **Sign in**: Your browser opens. Sign in to Spiral, pick the organization and click Authorize, then return to the terminal. No API key. Inside opencode, `/mcps` starts the same sign-in.

```bash
opencode mcp auth spiral
```

3. **Check it**: Spiral shows as connected.

```bash
opencode mcp list
```

### Use the Spiral skill

1. **Send opencode 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 Qwen2.5 72B 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/qwen-2.5-72b-instruct",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: opencode finds Qwen2.5 72B Instruct on Spiral, shows the price and runs it.

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

## Qwen2.5 72B Instruct tools and prices

- [Qwen2.5 72B Instruct](https://spiral.ai/tools/openrouter/qwen/qwen-2.5-72b-instruct) (OpenRouter): Up to $0.002147 per run at the default inputs. $0.468 per 1M input tokens, $0.52 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 Qwen2.5 72B Instruct in opencode?

Turn project code, specifications, and tables into implementation drafts, tests, and structured findings with Qwen2.5 72B Instruct. Give the agent a focused task and require checks against the project's expected behavior.

### Extract records from business text

Turn order descriptions, support messages, or meeting notes into JSON. Define required values, how to represent missing information, and which source passage supports each record.

### Explain tables

Read tables supplied as text and explain trends, inconsistent entries, and missing values. Ask for row references so a reviewer can check the findings.

### Localize help content

Draft Chinese, English, Spanish, French, or Japanese versions of support answers. Supply a terminology list and ask it to preserve product names, numbers, and policy conditions.

### Draft code and tests

Generate a function, explain an implementation, or propose tests from a written specification. Keep the request focused and run the resulting code against the expected behavior.

### Compare long documents

Compare supplied policy text, technical specifications, or research notes. Request a short comparison with supporting passages, and review conclusions drawn from long inputs.

### Prepare recurring operations briefs

Summarize supplied status updates into decisions, blockers, and next actions. Keep the same output structure each time and distinguish reported facts from unanswered questions.

### Develop technical explanations

Explain a mathematical method or programming concept for a defined audience. Request worked examples and explicit assumptions, then verify calculations or executable examples.

## Overview

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

## Qwen2.5 72B Instruct in other agents

- [Amp](https://spiral.ai/integrations/amp/qwen2-5-72b-instruct)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/qwen2-5-72b-instruct)
- [Claude](https://spiral.ai/integrations/claude-desktop/qwen2-5-72b-instruct)
- [Claude Code](https://spiral.ai/integrations/claude-code/qwen2-5-72b-instruct)
- [Cline](https://spiral.ai/integrations/cline/qwen2-5-72b-instruct)
- [Codex](https://spiral.ai/integrations/codex/qwen2-5-72b-instruct)
- [Cursor](https://spiral.ai/integrations/cursor/qwen2-5-72b-instruct)
- [Devin](https://spiral.ai/integrations/devin/qwen2-5-72b-instruct)
- [dots](https://spiral.ai/integrations/dots/qwen2-5-72b-instruct)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/qwen2-5-72b-instruct)
- [Gemini](https://spiral.ai/integrations/gemini/qwen2-5-72b-instruct)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/qwen2-5-72b-instruct)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/qwen2-5-72b-instruct)
- [goose](https://spiral.ai/integrations/goose/qwen2-5-72b-instruct)
- [Grok](https://spiral.ai/integrations/grok/qwen2-5-72b-instruct)
- [Grok Build](https://spiral.ai/integrations/grok-build/qwen2-5-72b-instruct)
- [Hermes](https://spiral.ai/integrations/hermes/qwen2-5-72b-instruct)
- [Instinct](https://spiral.ai/integrations/instinct/qwen2-5-72b-instruct)
- [Manus](https://spiral.ai/integrations/manus/qwen2-5-72b-instruct)
- [Muse](https://spiral.ai/integrations/muse/qwen2-5-72b-instruct)
- [Muse Code](https://spiral.ai/integrations/muse-code/qwen2-5-72b-instruct)
- [n8n](https://spiral.ai/integrations/n8n/qwen2-5-72b-instruct)
- [Notion](https://spiral.ai/integrations/notion/qwen2-5-72b-instruct)
- [OpenClaw](https://spiral.ai/integrations/openclaw/qwen2-5-72b-instruct)
- [Perplexity](https://spiral.ai/integrations/perplexity/qwen2-5-72b-instruct)
- [Raycast](https://spiral.ai/integrations/raycast/qwen2-5-72b-instruct)
- [VS Code](https://spiral.ai/integrations/vscode/qwen2-5-72b-instruct)
- [Windsurf](https://spiral.ai/integrations/windsurf/qwen2-5-72b-instruct)
- [Zapier](https://spiral.ai/integrations/zapier/qwen2-5-72b-instruct)

## More opencode guides

- [Qwen Plus](https://spiral.ai/integrations/opencode/qwen-plus)
- [Qwen2.5 Coder 32B Instruct](https://spiral.ai/integrations/opencode/qwen2-5-coder-32b-instruct)
- [Pixtral Large 2502](https://spiral.ai/integrations/opencode/pixtral-large-2502)
- [Qwen2.5-VL 72B Instruct](https://spiral.ai/integrations/opencode/qwen2-5-vl-72b-instruct)
- [Phi-4](https://spiral.ai/integrations/opencode/phi-4)
- [Qwen3 14B](https://spiral.ai/integrations/opencode/qwen3-14b)

## Questions about Qwen2.5 72B Instruct in opencode

**Can opencode use Qwen2.5 72B Instruct?**

Yes. Add Spiral to opencode as an MCP server, and opencode can find and run Qwen2.5 72B Instruct like any built-in tool.

**How much does Qwen2.5 72B Instruct cost in opencode?**

Up to $0.002147 per run at the default inputs. $0.468 per 1M input tokens, $0.52 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your opencode plan is separate.

**Do I need an OpenRouter account to use Qwen2.5 72B Instruct in opencode?**

No. Spiral runs Qwen2.5 72B 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 Qwen2.5 72B Instruct in opencode?**

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

**Does opencode wait while Qwen2.5 72B Instruct runs?**

Qwen2.5 72B Instruct runs in the background. opencode starts the run, checks on it and gives you the text, json when it's ready.

**What is Qwen2.5 72B Instruct good at?**

It is useful for tasks that combine writing with structured information: interpreting tables, extracting JSON, drafting multilingual text, and producing code. Qwen also reports improved math and coding results over Qwen2 72B Instruct.

**How much text can Qwen2.5 72B Instruct handle?**

Qwen documents a full context length of 131,072 tokens with long-context extension and an output length of 8,192 tokens. Treat those as the maker's model specifications; for a long report, request an outline and draft one section at a time.

**Which languages does Qwen2.5 support?**

Qwen lists more than 29 supported languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, and Japanese. For translation work, specify the target language, audience, and terms that must stay unchanged.

**Is Qwen2.5 72B Instruct useful for JSON extraction?**

Yes. Qwen identifies structured data understanding and JSON generation as areas of improvement. Give it an exact output shape and missing-value rules, then validate the result before using it downstream.

**How should I prompt Qwen2.5 72B Instruct?**

State the task, audience, source material, and required output format explicitly. Qwen's published sampling configuration uses temperature 0.7, top-p 0.8, and top-k 20; use it as a starting point and evaluate changes on representative examples.

