Feature Request: Add MiniMax AI Model Support (including M2.1-lightning)

Description

I would like to request adding MiniMax AI models to the supported model list. MiniMax is a leading Chinese AI company that provides powerful language models with competitive pricing and excellent performance. Adding support for these models would significantly increase the options available to OpenCode users, particularly for Chinese language support.

Requested Models

MiniMax (base model)
MiniMax M2.1-lightning (faster/lighter variant for quick tasks)

Use Case

I want to use MiniMax models for:

Coding tasks and programming assistance
General conversation and Q&A
AI-assisted development workflows
Chinese language content generation

Additional Information

MiniMax offers:

Competitive pricing compared to other providers
Strong performance for various tasks
Excellent Chinese language support
Fast inference speed (especially M2.1-lightning variant)

References

MiniMax Official Website: https://www.minimaxi.com/
MiniMax API Documentation: https://api.minimaxi.com/

Implementation Checklist

Labels

feature-request
provider-request

Thank you for considering this feature request!

This issue might be a duplicate of existing issues. Please check:

[FEATURE]: Please add the new GLM 4.7 and MiniMax M2.1 models to Nvidia. #7953: [FEATURE]: Please add the new GLM 4.7 and MiniMax M2.1 models to Nvidia. (Also requests MiniMax M2.1 model support)
MiniMax M2 <think> tags #4033: MiniMax M2 tags (Discusses MiniMax M2 model usage and rendering)
[FEATURE]: Add kimi-k2-thinking and minimax-m2 model support for OpenCode Zen provider #4190: [FEATURE]: Add kimi-k2-thinking and minimax-m2 model support for OpenCode Zen provider (Previously requested MiniMax M2 support)

Feel free to ignore if none of these address your specific case.

Important Clarification on Exact Model Names

Thank you for creating this issue! However, I need to provide more precise model names to avoid GitHub bot misclassification.

Exact Model Name for Minimax-cn Provider:

minimax-cn/M2.1-lightning (exact format with hyphen: M2.1-lightning, not "M2.1" or "M2_1")

This is a specific model variant from MiniMax that offers extremely fast inference speed. The model name follows the format: provider-name/model-name.

Key Distinctions:

This is NOT the base "MiniMax" or "MiniMax M2.1" model
The -lightning suffix is critical - it indicates the optimized fast variant
When users try to add this model manually to the minimax-cn provider configuration, it fails to communicate with the API

Additional Context:

The MiniMax M2.1-lightning model provides:

Significantly faster response times than the base M2.1 model
Excellent performance for coding tasks
Competitive pricing

Please update the implementation to include this exact model name to ensure proper API communication.

To stay organized issues are automatically closed after 90 days of no activity. If the issue is still relevant please open a new one.

Navigation Menu

Repository navigation

Feature Request: Add MiniMax AI Model Support (including M2.1-lightning) #7997