Inference provider · Inference gateway
Vercel AI Gateway
Verified Aug 2026
Quick answer: One Vercel key across hundreds of lab models at list price (0% markup), with failover and first-party wiring into Claude Code, Codex, OpenCode, and Cline.
Best for
One Vercel key across hundreds of lab models at list price (0% markup), with failover and first-party wiring into Claude Code, Codex, OpenCode, and Cline.
Skip if
You want Cloudflare’s own GPUs (Workers AI), you meant v0 the vibe builder, or you already like OpenRouter’s catalog UX.
Who it fits
- One Vercel key across hundreds of lab models at list price (0% markup), with failover and first-party wiring into Claude Code, Codex, OpenCode, and Cline.
- Builders who start in the browser without a local IDE setup
Pros
One key, hundreds of models (OpenAI, Anthropic, SpaceXAI, Gemini, OSS) including text, image, and video; tokens at provider list with 0% markup — including BYOK; automatic failover if a provider degrades — same model on another host; native setup for Claude Code, Codex, OpenCode, Cline, and Grok Build; drop-in OpenAI Chat Completions, Responses, and Anthropic Messages APIs.
Cons
Extra hop vs calling a lab directly — you still depend on Vercel routing; credits wallet by default — not a free always-on hobby pool; some spend/security controls sit on higher Vercel plan tiers; not v0 — this is the model gateway, not the vibe builder.
Facts
- Locality:
- Cloud-native
- Surfaces:
- Web
- Maturity:
- General availability
- BYOK:
- Bring your own API key
- Open source:
- Source available
- Verified:
- Aug 2026
Agent standards & memory
Compare one neighbor first
Vendor pages sell hard. Read one alternative on this site, then leave for the official link if the fit still holds.
Official links
Alternatives
Someone asks you about vibe-coding software? Share this site with them — we are adding more useful data regularly. agents.dancingteeth.net