HeadsUpAI

Vercel adds MiniMax M3 to AI Gateway for 1M context agentic workflows

Vercel has added MiniMax M3 to its AI Gateway, a unified API for calling models. The model is the first from MiniMax to feature a 1-million-token context window (the data volume a model processes at once) and native multimodality. It uses MiniMax Sparse Attention to maintain performance across massive inputs while optimized for multi-turn collaboration.
Platform
Vercel AI Gateway
Model Identifier
minimax/minimax-m3
Next.js Eval Score
96%
Promotion
50% off inference
Access
Vercel AI SDK

The integration supports long-horizon execution for repository-wide engineering and autonomous browsing. By hosting MiniMax M3 on the AI Gateway, Vercel removes the friction of managing separate provider accounts. This follows the OpenRouter integration and tracks alongside Cloudflare's MiniMax M3 integration, signaling rapid ecosystem adoption.

Developers can call minimax/minimax-m3 via the AI SDK for software fixes or multimodal analysis. Vercel is offering 50% off inference (running a model to generate output) for the first week. The model is highly effective for Next.js development, scoring 96% on platform evaluations when using agentic documentation.

View the full update on vercel.com
Vercel Developers
Vercel Developers
@vercel_dev
X

MiniMax M3 is available on AI Gateway. MiniMax's first long-context model, with support for multimodal inputs. 50% off for the next week. ๐š–๐š˜๐š๐šŽ๐š•: '๐š–๐š’๐š—๐š’๐š–๐šŠ๐šก/๐š–๐š’๐š—๐š’๐š–๐šŠ๐šก-๐š–๐Ÿน' https://t.co/Be8lceGhaU

8retweets69likes
View on X

Still wondering? A few quick answers below.

MiniMax M3 is a foundation model designed for long-context and multimodal tasks. It features a 1-million-token context window and is optimized for agentic workflows, such as software engineering and web browsing. It uses a specialized sparse attention architecture to maintain efficiency when processing large amounts of data across multiple turns.

Developers can access the model by setting the model parameter to minimax/minimax-m3 within the Vercel AI SDK. Because it is hosted on the Vercel AI Gateway, users do not need to create a separate MiniMax account or manage additional API keys to begin building with the model.

The Vercel AI Gateway provides a unified API for calling models while offering built-in tools for tracking usage, managing costs, and configuring retries. It reflects provider pricing with no platform markup and currently includes a 50% discount on MiniMax M3 inference for the first week of availability.

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards โ†’

Share this update