Vercel.AI.Gateway
Vercel AI Gateway provider.
Classes
Section titled “Classes”| Type | Summary |
|---|---|
| GatewayAuthenticationError | Authentication failed. |
| GatewayBatchApi | AI Gateway batch API. Downloads the JSONL results of a batch. |
| GatewayDecisionModel | Answers questions through the Gateway /decision-model route. |
| GatewayEmbeddingModel | Gateway embedding model. |
| GatewayError | Base class for AI Gateway errors. The message includes [generationId] when one was returned. |
| GatewayErrors | Parses Gateway auth headers and error bodies. |
| GatewayFailedDependencyError | A dependency required by the Gateway request was unavailable. Status 424 is not retried. |
| GatewayForbiddenError | The request was rejected by policy. |
| GatewayImageModel | Gateway image model. |
| GatewayInternalServerError | The Gateway failed internally. |
| GatewayInvalidRequestError | The Gateway rejected the request. |
| GatewayLanguageModel | Gateway language model. The request body is the V4 call options. |
| GatewayModelNotFoundError | The requested model is not available. |
| GatewayNotFoundError | A Gateway resource was not found. |
| GatewayOptions | AI Gateway settings. The default base URL is https://ai-gateway.vercel.sh/v4/ai. |
| GatewayProvider | Vercel AI Gateway provider. Language, embedding, and image calls use the Gateway V4 routes (/language-model, /embedding-model, /image-model). |
| GatewayRateLimitError | A Gateway rate limit. |
| GatewayResponseError | The Gateway error body could not be read. |
| GatewayServiceCollectionExtensions | Registers the Gateway provider. |
| GatewayTakoSearch | JSON schemas for the Gateway Tako search tool. |
| GatewayTimeoutError | The client timed out before the Gateway responded. |
| V4Json | Serializes call options and parses results in the Language Model V4 JSON shape the Gateway expects. |
Apache-2.0 · Copyright 2023 Vercel, Inc. · unofficial community port, not an official Vercel product