minimax-m2-5-best-practices

name: minimax-m2-5-best-practices description: MiniMax M2.5 best practices for effective prompting. Use when user mentions MiniMax, M2.5, or wants to optimize LLM prompts for better results. license: MIT compatibility: Claude Code, OpenCode, Cursor, Codex metadata: author: minimax-tools version: 1.0.0

MiniMax M2.5 Best Practices

Master effective prompting techniques for the MiniMax M2.5 model to get better results.

When to Use

Apply these practices when:

• User mentions MiniMax, M2.5, or minimax
• User wants to optimize LLM prompts
• User asks how to get better results from MiniMax
• User provides prompt feedback or wants improvement

Core Principles

1. Be Clear and Specific with Instructions

Pattern: [ACTION] + [CONTEXT] + [EXPECTED OUTPUT FORMAT]

Effective:

Create an enterprise-grade data visualization website. Integrate as many rich analytical features and interactive functions as possible, going beyond basic display formats to build a fully-featured digital solution.

Less Effective:

Create a visualization website

2. Explain Your Intent

Pattern: Tell the model "why" - context helps provide accurate answers.

Effective:

Your response will be read aloud by a text-to-speech model, so present it in plain text format and avoid using document symbol formatting.

Less Effective:

Do not use document symbols

3. Focus on Examples and Details

Pattern: Show what you want with good/bad examples.

Effective:

Please write a product description following this example:

[Good: This desk lamp uses full-spectrum LED technology that simulates natural morning light to gently wake you up. It features 6 brightness levels to meet your different needs for reading, working, and resting.]

Please avoid vague descriptions like this:

[Bad: This desk lamp is great, the light is comfortable, and the design is nice.]

Now write a description for a 'smart thermos'.

4. Long Task Reasoning

For complex, multi-step tasks, include:

This is a very lengthy task. It's recommended that you make full use of the complete output context to handle it—keep the total input and output tokens within 200k tokens. Make full use of the context window length to complete the task thoroughly and avoid exhausting tokens.

Quick Reference

Related Resources

For detailed templates and anti-patterns, see:

• references/prompt-templates.md - Ready-to-use prompt patterns
• references/anti-patterns.md - Common mistakes to avoid