Vercel's June 2026 "AI Gateway Production Index," published June 8, 2026 and covering May 2026 data, reports that total tokens routed through AI Gateway grew 20% month-over-month while total spend grew 43%, meaning customers paid roughly 20% more per token on average. The headline shift was DeepSeek, whose share of tok The report frames two broader trends for developers building on the gateway: low-cost models are now capable enough to enter production mixes and are reshaping the volume leaderboard, and teams are responding with more deliberate routing between cheap and frontier models rather than pulling back on spend. Vercel also c

Vercel published an "AI Gateway Production Index" blog on May 12, 2026, aggregating seven months of production traffic from over 200,000 teams routing tens of trillions of tokens through its AI Gateway. The report positions the Gateway as a unique vantage point on real application workloads and highlights that Anthropi The index adds that OpenAI's spend share roughly tripled from March to April following the GPT-5.4/5.5 releases, that high-volume workloads route across an average of 30+ distinct models, and that agentic workloads now account for 59% of all token volume—up 2x over six months. Vercel frames the divergence between spend

On July 31, 2026, Vercel shipped spend budgets for AI Gateway scoped to teams and projects, not just individual API keys, completing the cost-control story for agent builders. A budget attaches to one of three scopes — team, project, or API key — and a single request must pass every budget it falls under; if any scope Default budgets can be set for projects or API keys, and every scope without an explicit budget inherits the default automatically, while explicit budgets always override defaults. The refresh period is daily, weekly, monthly (the default), or none for a cumulative cap that never resets, and BYOK (bring your own key) s

Vercel announced a new framework and enterprise controls for agentic AI infrastructure at its Ship annual conference on June 17, 2026, broadening its platform beyond front-end hosting into deeper enterprise deployment of AI agents. The launch targets how enterprises deploy, run, and scale with AI agents, signaling Verc The new agentic AI infrastructure products extend Vercel's existing AI Gateway and Fluid Compute stack, giving enterprise teams more controls for production agent deployments. This positions Vercel more directly against other AI-serving infrastructure providers and reinforces the agentic application framework as a stra

xAI has officially released Grok 4.3, the latest iteration of its AI model, with Elon Musk announcing availability on Vercel today. The rollout follows a beta period that began April 17 for SuperGrok Heavy subscribers, and the model is now broadly accessible through xAI's API, Vercel AI Gateway, and OpenRouter.Source:

You can now access Qwen 3.7 Max on Vercel's AI Gateway with no markup and no other provider accounts required.

Vercel raised $9.3B at Series F, cementing its position as the AI cloud for developers. What the funding means for the platform roadmap and Next.js.

Vercel AI Gateway launches in August 2025, offering a unified platform for AI model integration, supporting hundreds of models to streamline development and man