MiniMax M3.1 Flash now available on AI Gateway for free
Vercel’s AI Gateway now offers MiniMax M3.1 Flash at no cost until October 22, providing a 1M-token context window for coding and agentic tasks before reverting to standard pricing.
MiniMax M3.1 Flash, an open-weight multimodal model from MiniMax, is accessible for free on Vercel’s AI Gateway until October 22. The model supports a 1M-token context window, enabling processing of long documents and extended agentic workflows. It is optimized for complex coding tasks, including frontend development, targeted bug fixes, and component testing, as well as multi-step agentic processes across applications.
To access the free tier, users must ensure MiniMax serves the request by sorting the provider in the gateway’s provider options. After October 22, requests will bill at MiniMax’s standard rate. The model can be called via AI SDK, Chat Completions, Responses, and Anthropic Messages APIs, or integrated with coding agents connected to AI Gateway.
Developers can invoke the model using the identifier minimax/minimax-m3.1-flash. For coding agents like Claude Code, Codex, and Cursor, the latest Vercel CLI must be installed, followed by running setup to select the model. Detailed instructions are available in the coding agents guide, and the model can also be tested in the AI Gateway model playground.
AI Gateway provides a unified API for model invocation, usage tracking, and cost management, with features such as retries, failover, and performance optimizations. It includes Zero Data Retention support, budget controls for API keys, and routing rules. The platform reflects provider pricing without markup or platform fees on inference, including Bring Your Own Key requests.