[Integrations](https://spiral.ai/integrations) › [n8n](https://spiral.ai/integrations/n8n) › [Hy4 Preview](https://spiral.ai/integrations/n8n/hy4-preview)

# How to Use Hy4 Preview in n8n

Hy4 Preview handles long-context coding, office analysis, and scientific research. Its output ceiling is 64,000 tokens versus Hy4's 65,536, and Tencent documents a tendency to over-check its work.

|  |  |
| --- | --- |
| Model | Hy4 Preview |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.013357 per run at the default inputs. $1.0842 per 1M input tokens, $3.2513 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, with a Spiral API key |

## Prompts to try with Hy4 Preview in n8n

- Each week, use Hy4 Preview to compare the supplied expense table with the stated budget rules. Report discrepancies and missing evidence without inventing explanations.
- Each evening, use Hy4 Preview to analyze the supplied test-run summaries. Cluster recurring failures and propose the next diagnostic step for each cluster.
- Each Friday, use Hy4 Preview with deep reasoning to review the supplied experiment results. Compare them with the stated hypothesis and return a brief assessment with one verification pass.
- Every Monday at 9am: ask Hy4 Preview through Spiral to review the diff on this branch and list the three riskiest changes.
- Get a second opinion from Hy4 Preview on this function, then tell me what the call cost.

## How to set up Hy4 Preview in n8n

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

### Use the Spiral skill

1. **Send n8n 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 Spiral's MCP server

1. **Create a key**: Create a key under API keys in the console.

2. **Add an MCP Client node**: In a workflow, add an MCP Client node (or, under an AI Agent node, an MCP Client Tool) and set:

```text
https://api.spiral.ai/mcp
```

3. **Pick a tool**: The node lists Spiral's tools. In an MCP Client node pick one, such as `run`, and give it its arguments:

```json
{
  "provider": "fal",
  "model": "fal-ai/flux/schnell",
  "input": { "prompt": "a lighthouse at dawn" },
  "wait": true
}
```

4. **Run Hy4 Preview in a workflow step**: Give the step an instruction like the ones below and run it on your trigger or schedule. Spiral finds Hy4 Preview, prices it and runs it each time the step fires.

### 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 Hy4 Preview**: 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": "tencent/hy4-preview",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: n8n finds Hy4 Preview on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/n8n

## Hy4 Preview tools and prices

- [Tencent: Hy4 preview](https://spiral.ai/tools/openrouter/tencent/hy4-preview) (OpenRouter): Up to $0.013357 per run at the default inputs. $1.0842 per 1M input tokens, $3.2513 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 Hy4 Preview in n8n?

Tencent Hy4 can help produce recurring analyses from newly supplied information. Define the reporting period, evidence requirements, and stopping point for each run.

### Fix a bug across several files

Trace a failing behavior through the relevant code, propose a focused patch, and check it against the reported failure.

### Plan a code migration

Compare existing code with the target requirements, identify dependencies, and break the migration into changes with acceptance checks.

### Turn working notes into a report

Combine supplied notes into a report with decisions, unresolved questions, and evidence attached to each conclusion.

### Review a financial model

Examine supplied assumptions and tables, explain the calculations, and flag inconsistencies for review.

### Prototype a small game

Build a playable loop from a written brief, then refine controls, rules, and failure states through testing.

### Develop a research computation

Translate a mathematical or scientific question into a computational approach with explicit assumptions and validation checks.

## Overview

- [OpenRouter](https://spiral.ai/providers/openrouter)
- [All n8n guides](https://spiral.ai/integrations/n8n)

## Hy4 Preview in other agents

- [Amp](https://spiral.ai/integrations/amp/hy4-preview)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/hy4-preview)
- [Claude](https://spiral.ai/integrations/claude-desktop/hy4-preview)
- [Claude Code](https://spiral.ai/integrations/claude-code/hy4-preview)
- [Cline](https://spiral.ai/integrations/cline/hy4-preview)
- [Codex](https://spiral.ai/integrations/codex/hy4-preview)
- [Cursor](https://spiral.ai/integrations/cursor/hy4-preview)
- [Devin](https://spiral.ai/integrations/devin/hy4-preview)
- [dots](https://spiral.ai/integrations/dots/hy4-preview)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/hy4-preview)
- [Gemini](https://spiral.ai/integrations/gemini/hy4-preview)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/hy4-preview)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/hy4-preview)
- [goose](https://spiral.ai/integrations/goose/hy4-preview)
- [Grok](https://spiral.ai/integrations/grok/hy4-preview)
- [Grok Build](https://spiral.ai/integrations/grok-build/hy4-preview)
- [Hermes](https://spiral.ai/integrations/hermes/hy4-preview)
- [Instinct](https://spiral.ai/integrations/instinct/hy4-preview)
- [Manus](https://spiral.ai/integrations/manus/hy4-preview)
- [Muse](https://spiral.ai/integrations/muse/hy4-preview)
- [Muse Code](https://spiral.ai/integrations/muse-code/hy4-preview)
- [Notion](https://spiral.ai/integrations/notion/hy4-preview)
- [OpenClaw](https://spiral.ai/integrations/openclaw/hy4-preview)
- [opencode](https://spiral.ai/integrations/opencode/hy4-preview)
- [Perplexity](https://spiral.ai/integrations/perplexity/hy4-preview)
- [Raycast](https://spiral.ai/integrations/raycast/hy4-preview)
- [VS Code](https://spiral.ai/integrations/vscode/hy4-preview)
- [Windsurf](https://spiral.ai/integrations/windsurf/hy4-preview)
- [Zapier](https://spiral.ai/integrations/zapier/hy4-preview)

## More n8n guides

- [Hy4 Flash](https://spiral.ai/integrations/n8n/hy4-flash)
- [Inkling](https://spiral.ai/integrations/n8n/inkling)
- [Hy4](https://spiral.ai/integrations/n8n/hy4)
- [Inkling 2 Mini](https://spiral.ai/integrations/n8n/inkling-2-mini)
- [Hy3](https://spiral.ai/integrations/n8n/hy3)
- [Inkling 2 Small](https://spiral.ai/integrations/n8n/inkling-2-small)

## Questions about Hy4 Preview in n8n

**Can n8n use Hy4 Preview?**

Yes. Add Spiral to n8n as an MCP server, and n8n can find and run Hy4 Preview like any built-in tool.

**How much does Hy4 Preview cost in n8n?**

Up to $0.013357 per run at the default inputs. $1.0842 per 1M input tokens, $3.2513 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your n8n plan is separate.

**Do I need an OpenRouter account to use Hy4 Preview in n8n?**

You need no OpenRouter account. n8n uses one Spiral API key for every tool, and Spiral runs Hy4 Preview on its own OpenRouter account.

**How long does it take to set up Hy4 Preview in n8n?**

About two minutes: create a key, then add an MCP Client node, then pick a tool.

**Does n8n wait while Hy4 Preview runs?**

Hy4 Preview runs in the background. n8n starts the run, checks on it and gives you the text, json when it's ready.

**What is Tencent Hy4 useful for?**

Coding, office analysis, and scientific research are documented uses. Give it work that requires connecting evidence and completing several related steps.

**How much context can Tencent Hy4 handle?**

Hy4 and Hy4 Preview each have a 1,048,576-token context window. For a large request, identify the relevant sections and state the exact question you need answered.

**How long can a Tencent Hy4 response be?**

Hy4 supports up to 65,536 completion tokens; Hy4 Preview supports up to 64,000. Specify the deliverable's length instead of requesting the maximum by default.

**Can Tencent Hy4 make factual mistakes?**

Yes. Tencent reported lower hallucination rates for Hy4 than Hy4 Preview in an internal evaluation, but errors remained. Ask for supporting passages and verify consequential claims against the original evidence.

**Can Tencent Hy4 help build a game?**

Game prototyping is a documented use. Start with a small playable loop, define the controls and win condition, and refine it after testing.

