Vercel AI Gateway
for LuumenAI
One gateway for model lookup, spend, and inference calls from the terminal
Connect Vercel AI Gateway and Luumen can read your model catalog, per-provider endpoint pricing and latency, credit balance, and spend reports on demand. When you want a call made — a chat completion, an embedding batch, an image generation — Luumen shows the model, the routing, and the request first, then waits for your approval.
The Vercel AI Gateway toolbox
22 tools: 9 read, 13 write. Reads answer instantly. Writes require approval by default. Everything is logged.
- ReadCount Message TokensEstimate input tokens for Anthropic-format messages, system content, and tools before sending a model request through Vercel AI Gateway.
- ReadGet Credit BalanceReturn the connected team's remaining AI Gateway credit balance and lifetime credit usage in USD.
- ReadGet GenerationLook up cost, token usage, provider routing, latency, and finish details for one completed Vercel AI Gateway generation.
- ReadGet ModelReturn detailed capabilities, modalities, limits, tags, and pricing for one exact Gateway model ID.
- ReadGet Model EndpointsCompare provider endpoints serving one Vercel AI Gateway model, including limits, pricing, caching, availability, throughput, and latency.
- ReadGet Spend ReportQuery aggregated Vercel AI Gateway spend and token usage.
- ReadList Model CatalogList AI Gateway model-to-provider routes and capability tags, with bounded local filtering by model text, provider, or tag.
- ReadList TypeSafe ModelsList evaluation models available through AI Gateway's TypeSafe-compatible API, including each accepted model name, description, and release date.
- ReadSearch ModelsSearch the AI Gateway model catalog by text, model type, and tags, returning capability and pricing metadata for matching models.
- WriteCompact ResponseCompact a long OpenAI Responses conversation into replayable context through Vercel AI Gateway. Approval by default
- WriteCreate Anthropic MessageCreate one non-streaming Anthropic Messages-compatible result through Vercel AI Gateway, with multimodal content, tools, extended thinking, and routing controls. Approval by default
- WriteCreate Chat CompletionCreate one non-streaming OpenAI Chat Completions-compatible result through Vercel AI Gateway, with multimodal messages, tools, structured output, fallback models, and routing controls. Approval by default
- WriteCreate EmbeddingsGenerate OpenAI-compatible vector embeddings through Vercel AI Gateway for text or token inputs, with optional dimensions and routing controls. Approval by default
- WriteEdit ImagesCreate edited images from up to 16 HTTPS or base64 data-URL source images and a text instruction. Approval by default
- WriteEvaluate StateEvaluate one shared state against one or more typed choice, score, or boolean questions using an AI Gateway evaluation model. Approval by default
- WriteGenerate ImagesGenerate one or more images from a text prompt through an image-only Vercel AI Gateway model. Approval by default
- WriteGenerate ResponseGenerate one non-streaming OpenAI Responses-compatible result through AI Gateway, with optional instructions, tools, structured output, reasoning, and routing controls. Approval by default
- WriteGenerate SpeechGenerate spoken audio from text through a Vercel AI Gateway speech model. Approval by default
- WriteRerank Object DocumentsRank structured object documents by relevance through the Cohere v1 route, using selected rank fields and optionally returning each source document. Approval by default
- WriteRerank Text DocumentsRerank string documents through the current Cohere v2 compatibility endpoint. Approval by default
- WriteTranscribe AudioTranscribe base64-encoded recorded audio through a Vercel AI Gateway transcription model. Approval by default
- WriteEvaluate with TypeSafe APIEvaluate one shared state through AI Gateway's TypeSafe-compatible System One API. Approval by default
One prompt, start to finish
What a governed Vercel AI Gateway run looks like inside Luumen.
Questions
Authorize once with API token. Luumen lists the scopes each action needs before you approve the connection, and credentials never appear in the chat.
Read actions answer immediately. Anything that writes — compact response, create anthropic message, create chat completion, create embeddings, and more — is shown as a plan and requires approval by default. Administrators configure that per tool, so you decide exactly which actions can ever run unattended.
You decide. Actions are granted per agent, skill, and team, and per environment — production is not staging. Read access can be broad while writes stay narrow.
Every call to Vercel AI Gateway — read or write, approved or declined — is recorded with the actor, the input, and the result, and can be linked to the ticket or change record.
Often connected alongside
All integrations
Adobe Marketing Agent MCPAdobe Experience Platform context, sandbox switches, and async task status
AffindaRead parsed documents, collections, and annotations — change them with approval
Agent MailGive the agent its own mailbox for reading threads and sending updates
AI/ML APIInspect assistants, threads, and runs — and change them as approved stepsPut Vercel AI Gateway to work with Luumen
Connect in minutes. Every action scoped, approved, and audited from day one.