GenAI PMVercel AI Gateway
Vercel’s model-routing and gateway product, mentioned in relation to spend and token share shifts across providers. Relevant to AI PMs for model traffic management and provider strategy.
Key Highlights
- Vercel AI Gateway is presented as a unified API for LLMs and vector databases with routing, caching, failover, and observability.
- Newsletter mentions emphasize its role in tracking and reacting to provider share shifts, including spend and token volume changes across Anthropic and OpenAI.
- Advanced budgets by key, team, and project make it relevant for AI PMs managing internal usage controls and cost accountability.
- The product is positioned for enterprise needs through zero data retention options, US/EU inference controls, and curated providers.
- Tailscale’s model router was described as using Vercel AI Gateway infrastructure, reinforcing its role as a backend control layer for AI traffic.
At a glance
- Website
- https://vercel.com/ai
- Category
- Gateway
Vercel AI Gateway
Overview
Vercel AI Gateway is Vercel’s model-routing and gateway product, described as a unified API for LLMs and vector databases with built-in routing, caching, failover, and real-time observability. Across newsletter mentions, it is positioned as infrastructure for managing model traffic across providers, controlling spend, improving uptime, and enforcing operational policies such as zero data retention and region-based inference.For AI Product Managers, Vercel AI Gateway matters because it sits at the control layer between application traffic and model providers. That makes it useful for provider strategy, cost governance, reliability, and experimentation. The newsletter repeatedly references shifts in provider spend and token share inside the gateway, showing how it can function not just as an integration layer but also as a decision surface for routing, budgeting, and adapting quickly as model pricing, performance, and availability change.
Key Developments
- 2026-07-06: Guillermo Rauch said Vercel AI Gateway used multiple token origins for Anthropic models to keep them online nearly 100% of the time, and added routing rules to automatically swap out retired models.
- 2026-07-18: Guillermo Rauch launched AI Gateway as Vercel’s unified API for LLMs and vector databases, highlighting built-in routing, caching, and real-time observability.
- 2026-08-02: Vercel introduced advanced spend budgets for AI Gateway, including budgets per key, team, and project, alongside failover, model/provider choice, and real-time observability.
- 2026-08-24: Guillermo Rauch said OpenAI Sol became Vercel’s fastest-growing frontier model after price reductions and discounts on Vercel AI Gateway, arguing gateways can capitalize on inference price volatility.
- 2026-09-04: Guillermo Rauch shared the `vercel ai-gateway coding-agents setup` command to point coding agents at AI Gateway, emphasizing uptime, observability, budgets, and easy model switching.
- 2026-09-12: Guillermo Rauch said Tailscale’s model router uses Vercel AI Gateway for zero data retention, zero markup, free BYOK, and cost/usage data, and framed AI gateways as infrastructure analogous to CDNs.
- 2026-09-13: Guillermo Rauch reiterated that Tailscale’s model router runs on Vercel AI Gateway infrastructure.
- 2026-09-17: Guillermo Rauch said new safety/reviewer behavior from `typesafeai` was coming to Vercel AI Gateway and would likely become the new default.
- 2026-09-21: Guillermo Rauch said Vercel AI Gateway supports zero data remotion (ZDR) and US- or EU-based inference, with curated providers and enterprise enforcement through the control plane.
- 2026-09-25: Guillermo Rauch shared a two-month recap of provider share shifts on Vercel AI Gateway: Anthropic remained #1 in spend but dropped from 69% to 40%, OpenAI rose from 10% to 24% and led in token count, Kimi K3 and DeepSeek captured about half of Anthropic’s loss, Opus 5.5 reached 10% of spend in two days, and OpenAI generated 62% of images.
Relevance to AI PMs
- Provider strategy and routing: AI PMs can use a gateway like this to switch providers or models without rewriting app logic, which helps when price, latency, availability, or quality shifts quickly.
- Budget and margin control: Features like budgets by key, team, and project make it easier to operationalize cost controls, allocate usage internally, and respond to pricing changes across providers.
- Reliability and policy enforcement: Built-in failover, routing rules, uptime protections, zero data retention options, and region-based inference help PMs translate enterprise requirements into production controls.
Related
- Vercel: Parent company behind AI Gateway and the broader platform context.
- Guillermo Rauch: Vercel CEO and the main public source of product updates and usage-share commentary referenced here.
- AI SDK / qexexeai-sdk-api / npx-ai-cli: Related developer tooling and interfaces that can connect applications and agents to Vercel AI Gateway.
- vercel-ai-wallet: Related to spend and usage management themes around AI infrastructure.
- Tailscale: Its model router was said to use Vercel AI Gateway as underlying infrastructure.
- typesafeai: Its safety-reviewer workflow was said to be coming to AI Gateway and potentially becoming the default.
- Anthropic, OpenAI, xAI: Key model providers discussed in relation to spend share, token volume, pricing, and routing decisions within the gateway.
- OpenAI Sol, Gemini 3 Pro, Claude Code, nano-banana-2, grok-imagine-video, grok-imagine-image, generatevideo-api: Examples of adjacent model, agent, and media-generation entities relevant to multi-provider traffic management.
- Stripe: Relevant by analogy to control-plane, usage, and billing infrastructure patterns for API-mediated services.
Frequently asked questions
What is Vercel AI Gateway?
Vercel’s model-routing and gateway product, mentioned in relation to spend and token share shifts across providers. Relevant to AI PMs for model traffic management and provider strategy.
What should AI product managers know about Vercel AI Gateway?
Vercel AI Gateway is presented as a unified API for LLMs and vector databases with routing, caching, failover, and observability. Newsletter mentions emphasize its role in tracking and reacting to provider share shifts, including spend and token volume changes across Anthropic and OpenAI. Advanced budgets by key, team, and project make it relevant for AI PMs managing internal usage controls and cost accountability.
Explore this topic
Newsletter Mentions (20)
“Guillermo Rauch shared an attached video recapping the last 2 months on Vercel AI Gateway: Anthropic remained #1 in spend but fell from 69% to 40%, while OpenAI rose from 10% to 24% and led in token count.”
#14 Guillermo Rauch shared an attached video recapping the last 2 months on Vercel AI Gateway: Anthropic remained #1 in spend but fell from 69% to 40%, while OpenAI rose from 10% to 24% and led in token count. Kimi K3 and DeepSeek accounted for ~half of Anthropic’s loss, Opus 5.5 reached 10% of spend in 2 days, and OpenAI accounted for 62% of image generations.
“Guillermo Rauch commented that Vercel AI Gateway offers zero data remotion (ZDR) and US- or EU-based inference.”
Guillermo Rauch commented that Vercel AI Gateway offers zero data remotion (ZDR) and US- or EU-based inference. He added that providers are curated and enterprises can enforce these requirements through the control plane.
“He added that it is coming to Vercel AI Gateway and will likely become the new default.”
#10 𝕏 Guillermo Rauch announced results from @typesafeai: fx defaults to auto mode, with a GPT Luna safety reviewer analyzing every command, while Jev is reportedly up to 18x faster at p95 and more accurate. He added that it is coming to Vercel AI Gateway and will likely become the new default.
“Guillermo Rauch shared that Tailscale’s model router uses Vercel AI Gateway as its underlying infrastructure.”
#7 in Guillermo Rauch shared that Tailscale’s model router uses Vercel AI Gateway as its underlying infrastructure.
“Guillermo Rauch commented that Tailscale’s model router uses Vercel AI Gateway for zero data retention, zero markup, free BYOK, and cost and usage data.”
#10 𝕏 Guillermo Rauch commented that Tailscale’s model router uses Vercel AI Gateway for zero data retention, zero markup, free BYOK, and cost and usage data. He described AI gateways as the new CDNs, calling direct-to-origin brittle and DIY alternatives painful and costly. #11 𝕏 Guillermo Rauch said Eve is a model-agnostic harness and shared ai-sdk.dev’s Harness Agent for abstracting harnesses.
“Guillermo Rauch shared the `vercel ai-gateway coding-agents setup` command, which points all coding agents to AI Gateway. He says it provides 100% uptime, observability, budgets, and easy switching.”
Guillermo Rauch shared the `vercel ai-gateway coding-agents setup` command, which points all coding agents to AI Gateway. He says it provides 100% uptime, observability, budgets, and easy switching.
“Guillermo Rauch commented that OpenAI Sol’s price reductions and discounts on Vercel AI Gateway made it Vercel’s fastest-growing frontier model, arguing that usage rises rapidly as inference costs fall.”
#2 𝕏 Guillermo Rauch commented that OpenAI Sol’s price reductions and discounts on Vercel AI Gateway made it Vercel’s fastest-growing frontier model, arguing that usage rises rapidly as inference costs fall. He said gateways can take advantage of price volatility to lower operating costs and increase margins, describing gateways as inevitable. #3 𝕏 Guillermo Rauch commented that fx’s extension philosophy centers on open protocols—MCP, Skills, Plugins, and Unix-style composability. He said libfx enables embedding fx into more complex programs and that users should be able to build their own CLI, background agent, or software factory for local or cloud use.
“in Guillermo Rauch announced advanced spend budgets for Vercel AI Gateway, including budgets per key, team, and project. Vercel’s AI Gateway also offers failover, model and provider choice, and realtime observability.”
#3 in Guillermo Rauch announced advanced spend budgets for Vercel AI Gateway, including budgets per key, team, and project. Vercel’s AI Gateway also offers failover, model and provider choice, and realtime observability.
“Guillermo Rauch launched ▲ AI Gateway, Vercel’s new unified API for LLMs and vector databases, offering built-in routing, caching, and real-time observability.”
#10 𝕏 Guillermo Rauch launched ▲ AI Gateway, Vercel’s new unified API for LLMs and vector databases, offering built-in routing, caching, and real-time observability.
“Guillermo Rauch built multiple token origins for Anthropic’s models in Vercel AI Gateway to keep them online nearly 100% of the time, and rolled out routing rules to automatically swap out any retired model.”
How Vercel AI Gateway keeps Anthropic models online #1 𝕏 Guillermo Rauch built multiple token origins for Anthropic’s models in Vercel AI Gateway to keep them online nearly 100% of the time, and rolled out routing rules to automatically swap out any retired model.
Related
An AI company mentioned for releasing a report on model behaviors and for internal tool use in later context. Relevant for PMs tracking safety reporting and frontier model operations.
Anthropic’s coding agent/environment used here to control physical tools in a long-running home experiment.
OpenAI is an AI company building foundation models and consumer products like ChatGPT. In this newsletter it is discussed in connection with GPT-6, Dots, ChatGPT for Teens, and College Planner.
Founder and leader at Vercel, cited for data on bot traffic and agentic usage. His comments reflect how websites and docs are increasingly consumed by agents rather than humans.
A developer platform whose network traffic and deployments are increasingly agentic according to the newsletter. It is also optimizing documentation content for agents.
A creator and product thinker sharing AI usage tracking and product documentation workflows. Relevant to PMs focused on measurement and prompt-to-spec workflows.
Elon Musk's AI company, referenced here through a bot that reviews websites and flags design issues. For PMs, it implies automated QA and product critique workflows.
A collaborative AI product design/building tool that now supports team visibility and shared workspaces. Useful for PMs collaborating on prototype and product workflows.
A payments company referenced for internal agent use and operational safeguards around agentic workflows.
An image-generation model or capability used in Google Earth on the web. It enables prompt-based visual reimagining of locations with satellite and 3D imagery.
Vercel’s SDK for integrating AI features into apps. The newsletter highlights token savings from a single line of code in DeepSeek-powered workflows.
A Gemini model variant used in a real workflow library project. The newsletter mentions it as one of the tools used to build the ChatPRD index.
Stay updated on Vercel AI Gateway
Get curated AI PM insights delivered daily — covering this and 1,000+ other sources.
Subscribe Free