[Integrations](https://spiral.ai/integrations) › [Instinct](https://spiral.ai/integrations/instinct) › [MiniMax M2.5](https://spiral.ai/integrations/instinct/minimax-m2-5)

# How to Use MiniMax M2.5 in Instinct

MiniMax M2.5 handles code changes, research with tools, and office drafts. Compared with MiniMax M2.1, it plans features, structure, and UI design more explicitly before writing code.

|  |  |
| --- | --- |
| Model | MiniMax M2.5 |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.005072 per run at the default inputs. $0.351 per 1M input tokens, $1.235 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, with a Spiral API key |

## Prompts to try with MiniMax M2.5 in Instinct

- Every Monday, use MiniMax M2.5 to check the official Python, TypeScript, and Go release notes for updates from the previous week. Produce a brief with dates, source links, and changes that could affect existing applications.
- Each morning, use MiniMax M2.5 to review new public issues in the repositories I follow through available tools. Group reproducible bug reports, summarize reproduction steps, and identify missing diagnostic information.
- Every Friday, use MiniMax M2.5 to research new tool-lending programs announced by public libraries that week. Prefer official announcements, include dates and links, and report when no verified announcements are found.
- At the end of each week, use MiniMax M2.5 to summarize the support conversations available to this agent. List recurring problems, supporting examples, and suggested investigation steps in no more than 600 words.
- On the first business day of each month, use MiniMax M2.5 to turn the previous month's research briefs into an eight-slide presentation outline. Include source links, changes since the prior month, and unresolved questions.

## How to set up MiniMax M2.5 in Instinct

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

### Use the Spiral skill

1. **Send Instinct 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 an API key**: In the console, open API keys and create a key for Instinct. Copy it.

2. **Send Instinct this message**: Instinct reads Spiral's skill and asks for your API key.

```text
set up https://spiral.ai/SKILL.md
```

3. **Give Instinct the key**: When Instinct asks for the key, add it where Instinct keeps secrets, never in the chat itself.

### 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 MiniMax M2.5**: 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": "minimax/minimax-m2.5",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Instinct finds MiniMax M2.5 on Spiral, shows the price and runs it.

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

## MiniMax M2.5 tools and prices

- [MiniMax: MiniMax M2.5](https://spiral.ai/tools/openrouter/minimax/minimax-m2.5) (OpenRouter): Up to $0.040482 per run at the default inputs. $0.351 per 1M input tokens, $1.235 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 MiniMax M2.5 in Instinct?

MiniMax M2.5 can turn recurring research and operational updates into reviewable briefs. Define the sources, comparison period, and evidence needed for each run.

### Fix a reproducible bug

Supply the failing behavior, relevant code, and test output. Ask for a cause, a focused patch, and a regression test.

### Turn a feature request into a specification

Define expected behavior, edge cases, interfaces, and acceptance tests before requesting implementation.

### Review a code change

Review a diff for correctness and missing tests. Request a concrete failure scenario for each finding.

### Port code between languages

Translate a Python utility into Go or TypeScript while preserving behavior. Include examples and tests to check the result.

### Build a sourced research brief

Use available search tools to investigate a narrow question, compare primary sources, and separate findings from unresolved claims.

### Draft spreadsheet formulas

Describe the columns and calculation rules. Request Excel formulas, assumptions, and sample cases with expected results.

### Prepare presentation content

Turn supplied evidence into a PowerPoint outline with slide titles, supporting points, speaker notes, and source references.

### Draft a structured report

Provide source text and a required outline. Request a Word-ready draft with headings, evidence, recommendations, and open questions.

## Overview

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

## MiniMax M2.5 in other agents

- [Amp](https://spiral.ai/integrations/amp/minimax-m2-5)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/minimax-m2-5)
- [Claude](https://spiral.ai/integrations/claude-desktop/minimax-m2-5)
- [Claude Code](https://spiral.ai/integrations/claude-code/minimax-m2-5)
- [Cline](https://spiral.ai/integrations/cline/minimax-m2-5)
- [Codex](https://spiral.ai/integrations/codex/minimax-m2-5)
- [Cursor](https://spiral.ai/integrations/cursor/minimax-m2-5)
- [Devin](https://spiral.ai/integrations/devin/minimax-m2-5)
- [dots](https://spiral.ai/integrations/dots/minimax-m2-5)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/minimax-m2-5)
- [Gemini](https://spiral.ai/integrations/gemini/minimax-m2-5)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/minimax-m2-5)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/minimax-m2-5)
- [goose](https://spiral.ai/integrations/goose/minimax-m2-5)
- [Grok](https://spiral.ai/integrations/grok/minimax-m2-5)
- [Grok Build](https://spiral.ai/integrations/grok-build/minimax-m2-5)
- [Hermes](https://spiral.ai/integrations/hermes/minimax-m2-5)
- [Manus](https://spiral.ai/integrations/manus/minimax-m2-5)
- [Muse](https://spiral.ai/integrations/muse/minimax-m2-5)
- [Muse Code](https://spiral.ai/integrations/muse-code/minimax-m2-5)
- [n8n](https://spiral.ai/integrations/n8n/minimax-m2-5)
- [Notion](https://spiral.ai/integrations/notion/minimax-m2-5)
- [OpenClaw](https://spiral.ai/integrations/openclaw/minimax-m2-5)
- [opencode](https://spiral.ai/integrations/opencode/minimax-m2-5)
- [Perplexity](https://spiral.ai/integrations/perplexity/minimax-m2-5)
- [Raycast](https://spiral.ai/integrations/raycast/minimax-m2-5)
- [VS Code](https://spiral.ai/integrations/vscode/minimax-m2-5)
- [Windsurf](https://spiral.ai/integrations/windsurf/minimax-m2-5)
- [Zapier](https://spiral.ai/integrations/zapier/minimax-m2-5)

## More Instinct guides

- [MiMo-V2.6 Flash](https://spiral.ai/integrations/instinct/mimo-v2-6-flash)
- [MiniMax M2.7](https://spiral.ai/integrations/instinct/minimax-m2-7)
- [MiMo-V2.5-Pro](https://spiral.ai/integrations/instinct/mimo-v2-5-pro)
- [MiniMax M3](https://spiral.ai/integrations/instinct/minimax-m3)
- [MiMo-V2.5](https://spiral.ai/integrations/instinct/mimo-v2-5)
- [MiniMax M3 (batch)](https://spiral.ai/integrations/instinct/minimax-m3-batch)

## Questions about MiniMax M2.5 in Instinct

**Can Instinct use MiniMax M2.5?**

Yes. Add Spiral to Instinct as an MCP server, and Instinct can find and run MiniMax M2.5 like any built-in tool.

**How much does MiniMax M2.5 cost in Instinct?**

Up to $0.005072 per run at the default inputs. $0.351 per 1M input tokens, $1.235 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Instinct plan is separate.

**Do I need an OpenRouter account to use MiniMax M2.5 in Instinct?**

You need no OpenRouter account. Instinct uses one Spiral API key for every tool, and Spiral runs MiniMax M2.5 on its own OpenRouter account.

**How long does it take to set up MiniMax M2.5 in Instinct?**

About two minutes: create an API key, then send Instinct this message, then give Instinct the key.

**Does Instinct wait while MiniMax M2.5 runs?**

MiniMax M2.5 runs in the background. Instinct starts the run, checks on it and gives you the text, json when it's ready.

**What is MiniMax M2.5 good at?**

Its documented strengths are coding, research with tools, and office work. Useful starting tasks include fixing a bug, comparing sources, or drafting a report with clear requirements.

**Which programming languages does MiniMax M2.5 cover?**

MiniMax lists Go, C, C++, TypeScript, Rust, Kotlin, Python, Java, JavaScript, PHP, Lua, Dart, and Ruby in its training coverage. Specify your language version, dependencies, and test command when requesting code.

**How much context is MiniMax M2.5 designed for?**

The published checkpoint is configured for 196,608 token positions. Treat that as a checkpoint specification; keep the supplied material focused and leave room for the response.

**Which generation settings should I start with?**

Start with MiniMax's published settings: temperature 1.0, top-p 0.95, and top-k 40. Compare changes against a fixed set of representative tasks before adopting them.

**Where does MiniMax M2.5 need extra checking?**

Difficult questions answered without tools deserve extra checking: MiniMax reports 19.4 on Humanity's Last Exam without tools, below M2.1's 22.2. Ask for evidence and check the supporting sources before relying on a difficult factual answer.

