@ai-sdk/gateway 4.0.0-beta.1 → 4.0.0-beta.108

package/dist/index.d.ts

@@ -1,11 +1,74 @@
-import { LanguageModelV3, ProviderV3, EmbeddingModelV3, ImageModelV3, Experimental_VideoModelV3, TypeValidationError } from '@ai-sdk/provider';
+import { LanguageModelV4, ProviderV4, EmbeddingModelV4, ImageModelV4, Experimental_VideoModelV4, RerankingModelV4, SpeechModelV4, TranscriptionModelV4, Experimental_RealtimeFactoryV4, TypeValidationError } from '@ai-sdk/provider';
 import * as _ai_sdk_provider_utils from '@ai-sdk/provider-utils';
 import { FetchFunction, InferSchema } from '@ai-sdk/provider-utils';
 
-type GatewayModelId = 'alibaba/qwen-3-14b' | 'alibaba/qwen-3-235b' | 'alibaba/qwen-3-30b' | 'alibaba/qwen-3-32b' | 'alibaba/qwen3-235b-a22b-thinking' | 'alibaba/qwen3-coder' | 'alibaba/qwen3-coder-30b-a3b' | 'alibaba/qwen3-coder-next' | 'alibaba/qwen3-coder-plus' | 'alibaba/qwen3-max' | 'alibaba/qwen3-max-preview' | 'alibaba/qwen3-max-thinking' | 'alibaba/qwen3-next-80b-a3b-instruct' | 'alibaba/qwen3-next-80b-a3b-thinking' | 'alibaba/qwen3-vl-instruct' | 'alibaba/qwen3-vl-thinking' | 'alibaba/qwen3.5-flash' | 'alibaba/qwen3.5-plus' | 'amazon/nova-2-lite' | 'amazon/nova-lite' | 'amazon/nova-micro' | 'amazon/nova-pro' | 'anthropic/claude-3-haiku' | 'anthropic/claude-3-opus' | 'anthropic/claude-3.5-haiku' | 'anthropic/claude-3.5-sonnet' | 'anthropic/claude-3.5-sonnet-20240620' | 'anthropic/claude-3.7-sonnet' | 'anthropic/claude-haiku-4.5' | 'anthropic/claude-opus-4' | 'anthropic/claude-opus-4.1' | 'anthropic/claude-opus-4.5' | 'anthropic/claude-opus-4.6' | 'anthropic/claude-sonnet-4' | 'anthropic/claude-sonnet-4.5' | 'anthropic/claude-sonnet-4.6' | 'arcee-ai/trinity-large-preview' | 'arcee-ai/trinity-mini' | 'bytedance/seed-1.6' | 'bytedance/seed-1.8' | 'cohere/command-a' | 'deepseek/deepseek-r1' | 'deepseek/deepseek-v3' | 'deepseek/deepseek-v3.1' | 'deepseek/deepseek-v3.1-terminus' | 'deepseek/deepseek-v3.2' | 'deepseek/deepseek-v3.2-thinking' | 'google/gemini-2.0-flash' | 'google/gemini-2.0-flash-lite' | 'google/gemini-2.5-flash' | 'google/gemini-2.5-flash-image' | 'google/gemini-2.5-flash-lite' | 'google/gemini-2.5-flash-lite-preview-09-2025' | 'google/gemini-2.5-flash-preview-09-2025' | 'google/gemini-2.5-pro' | 'google/gemini-3-flash' | 'google/gemini-3-pro-image' | 'google/gemini-3-pro-preview' | 'google/gemini-3.1-flash-image-preview' | 'google/gemini-3.1-flash-lite-preview' | 'google/gemini-3.1-pro-preview' | 'inception/mercury-2' | 'inception/mercury-coder-small' | 'kwaipilot/kat-coder-pro-v1' | 'meituan/longcat-flash-chat' | 'meituan/longcat-flash-thinking' | 'meta/llama-3.1-70b' | 'meta/llama-3.1-8b' | 'meta/llama-3.2-11b' | 'meta/llama-3.2-1b' | 'meta/llama-3.2-3b' | 'meta/llama-3.2-90b' | 'meta/llama-3.3-70b' | 'meta/llama-4-maverick' | 'meta/llama-4-scout' | 'minimax/minimax-m2' | 'minimax/minimax-m2.1' | 'minimax/minimax-m2.1-lightning' | 'minimax/minimax-m2.5' | 'mistral/codestral' | 'mistral/devstral-2' | 'mistral/devstral-small' | 'mistral/devstral-small-2' | 'mistral/magistral-medium' | 'mistral/magistral-small' | 'mistral/ministral-14b' | 'mistral/ministral-3b' | 'mistral/ministral-8b' | 'mistral/mistral-large-3' | 'mistral/mistral-medium' | 'mistral/mistral-nemo' | 'mistral/mistral-small' | 'mistral/mixtral-8x22b-instruct' | 'mistral/pixtral-12b' | 'mistral/pixtral-large' | 'moonshotai/kimi-k2' | 'moonshotai/kimi-k2-0905' | 'moonshotai/kimi-k2-thinking' | 'moonshotai/kimi-k2-thinking-turbo' | 'moonshotai/kimi-k2-turbo' | 'moonshotai/kimi-k2.5' | 'morph/morph-v3-fast' | 'morph/morph-v3-large' | 'nvidia/nemotron-3-nano-30b-a3b' | 'nvidia/nemotron-nano-12b-v2-vl' | 'nvidia/nemotron-nano-9b-v2' | 'openai/codex-mini' | 'openai/gpt-3.5-turbo' | 'openai/gpt-3.5-turbo-instruct' | 'openai/gpt-4-turbo' | 'openai/gpt-4.1' | 'openai/gpt-4.1-mini' | 'openai/gpt-4.1-nano' | 'openai/gpt-4o' | 'openai/gpt-4o-mini' | 'openai/gpt-4o-mini-search-preview' | 'openai/gpt-5' | 'openai/gpt-5-chat' | 'openai/gpt-5-codex' | 'openai/gpt-5-mini' | 'openai/gpt-5-nano' | 'openai/gpt-5-pro' | 'openai/gpt-5.1-codex' | 'openai/gpt-5.1-codex-max' | 'openai/gpt-5.1-codex-mini' | 'openai/gpt-5.1-instant' | 'openai/gpt-5.1-thinking' | 'openai/gpt-5.2' | 'openai/gpt-5.2-chat' | 'openai/gpt-5.2-codex' | 'openai/gpt-5.2-pro' | 'openai/gpt-5.4' | 'openai/gpt-5.4-2026-03-05' | 'openai/gpt-5.4-pro' | 'openai/gpt-5.4-pro-2026-03-05' | 'openai/gpt-5.3-chat' | 'openai/gpt-5.3-codex' | 'openai/gpt-oss-120b' | 'openai/gpt-oss-20b' | 'openai/gpt-oss-safeguard-20b' | 'openai/o1' | 'openai/o3' | 'openai/o3-deep-research' | 'openai/o3-mini' | 'openai/o3-pro' | 'openai/o4-mini' | 'perplexity/sonar' | 'perplexity/sonar-pro' | 'perplexity/sonar-reasoning' | 'perplexity/sonar-reasoning-pro' | 'prime-intellect/intellect-3' | 'vercel/v0-1.0-md' | 'vercel/v0-1.5-md' | 'xai/grok-2-vision' | 'xai/grok-3' | 'xai/grok-3-fast' | 'xai/grok-3-mini' | 'xai/grok-3-mini-fast' | 'xai/grok-4' | 'xai/grok-4-fast-non-reasoning' | 'xai/grok-4-fast-reasoning' | 'xai/grok-4.1-fast-non-reasoning' | 'xai/grok-4.1-fast-reasoning' | 'xai/grok-code-fast-1' | 'xiaomi/mimo-v2-flash' | 'zai/glm-4.5' | 'zai/glm-4.5-air' | 'zai/glm-4.5v' | 'zai/glm-4.6' | 'zai/glm-4.6v' | 'zai/glm-4.6v-flash' | 'zai/glm-4.7' | 'zai/glm-4.7-flashx' | 'zai/glm-5' | (string & {});
+type GatewayModelId = 'alibaba/qwen-3-14b' | 'alibaba/qwen-3-235b' | 'alibaba/qwen-3-30b' | 'alibaba/qwen-3-32b' | 'alibaba/qwen-3.6-max-preview' | 'alibaba/qwen3-235b-a22b-thinking' | 'alibaba/qwen3-coder' | 'alibaba/qwen3-coder-30b-a3b' | 'alibaba/qwen3-coder-next' | 'alibaba/qwen3-coder-plus' | 'alibaba/qwen3-max' | 'alibaba/qwen3-max-preview' | 'alibaba/qwen3-max-thinking' | 'alibaba/qwen3-next-80b-a3b-instruct' | 'alibaba/qwen3-next-80b-a3b-thinking' | 'alibaba/qwen3-vl-235b-a22b-instruct' | 'alibaba/qwen3-vl-instruct' | 'alibaba/qwen3-vl-thinking' | 'alibaba/qwen3.5-flash' | 'alibaba/qwen3.5-plus' | 'alibaba/qwen3.6-27b' | 'alibaba/qwen3.6-plus' | 'alibaba/qwen3.7-max' | 'alibaba/qwen3.7-plus' | 'amazon/nova-2-lite' | 'amazon/nova-lite' | 'amazon/nova-micro' | 'amazon/nova-pro' | 'anthropic/claude-3-haiku' | 'anthropic/claude-3.5-haiku' | 'anthropic/claude-haiku-4.5' | 'anthropic/claude-opus-4' | 'anthropic/claude-opus-4.1' | 'anthropic/claude-opus-4.5' | 'anthropic/claude-opus-4.6' | 'anthropic/claude-opus-4.7' | 'anthropic/claude-opus-4.8' | 'anthropic/claude-sonnet-4' | 'anthropic/claude-sonnet-4.5' | 'anthropic/claude-sonnet-4.6' | 'arcee-ai/trinity-large-preview' | 'arcee-ai/trinity-large-thinking' | 'arcee-ai/trinity-mini' | 'bytedance/seed-1.6' | 'bytedance/seed-1.8' | 'cohere/command-a' | 'deepseek/deepseek-r1' | 'deepseek/deepseek-v3' | 'deepseek/deepseek-v3.1' | 'deepseek/deepseek-v3.1-terminus' | 'deepseek/deepseek-v3.2' | 'deepseek/deepseek-v3.2-thinking' | 'deepseek/deepseek-v4-flash' | 'deepseek/deepseek-v4-pro' | 'google/gemini-2.5-flash' | 'google/gemini-2.5-flash-image' | 'google/gemini-2.5-flash-lite' | 'google/gemini-2.5-pro' | 'google/gemini-3-flash' | 'google/gemini-3-pro-image' | 'google/gemini-3-pro-preview' | 'google/gemini-3.1-flash-image' | 'google/gemini-3.1-flash-image-preview' | 'google/gemini-3.1-flash-lite' | 'google/gemini-3.1-flash-lite-preview' | 'google/gemini-3.1-pro-preview' | 'google/gemini-3.5-flash' | 'google/gemma-4-26b-a4b-it' | 'google/gemma-4-31b-it' | 'inception/mercury-2' | 'inception/mercury-coder-small' | 'interfaze/interfaze-beta' | 'kwaipilot/kat-coder-pro-v1' | 'kwaipilot/kat-coder-pro-v2' | 'meituan/longcat-flash-chat' | 'meituan/longcat-flash-thinking-2601' | 'meta/llama-3.1-70b' | 'meta/llama-3.1-8b' | 'meta/llama-3.2-11b' | 'meta/llama-3.2-1b' | 'meta/llama-3.2-3b' | 'meta/llama-3.2-90b' | 'meta/llama-3.3-70b' | 'meta/llama-4-maverick' | 'meta/llama-4-scout' | 'minimax/minimax-m2' | 'minimax/minimax-m2.1' | 'minimax/minimax-m2.1-lightning' | 'minimax/minimax-m2.5' | 'minimax/minimax-m2.5-highspeed' | 'minimax/minimax-m2.7' | 'minimax/minimax-m2.7-highspeed' | 'minimax/minimax-m3' | 'mistral/codestral' | 'mistral/devstral-2' | 'mistral/devstral-small' | 'mistral/devstral-small-2' | 'mistral/magistral-medium' | 'mistral/magistral-small' | 'mistral/ministral-14b' | 'mistral/ministral-3b' | 'mistral/ministral-8b' | 'mistral/mistral-large-3' | 'mistral/mistral-medium' | 'mistral/mistral-medium-3.5' | 'mistral/mistral-nemo' | 'mistral/mistral-small' | 'mistral/pixtral-12b' | 'mistral/pixtral-large' | 'moonshotai/kimi-k2' | 'moonshotai/kimi-k2-thinking' | 'moonshotai/kimi-k2.5' | 'moonshotai/kimi-k2.6' | 'moonshotai/kimi-k2.7-code' | 'moonshotai/kimi-k2.7-code-highspeed' | 'morph/morph-v3-fast' | 'morph/morph-v3-large' | 'nvidia/nemotron-3-nano-30b-a3b' | 'nvidia/nemotron-3-super-120b-a12b' | 'nvidia/nemotron-3-ultra-550b-a55b' | 'nvidia/nemotron-nano-12b-v2-vl' | 'nvidia/nemotron-nano-9b-v2' | 'openai/gpt-3.5-turbo' | 'openai/gpt-3.5-turbo-instruct' | 'openai/gpt-4-turbo' | 'openai/gpt-4.1' | 'openai/gpt-4.1-mini' | 'openai/gpt-4.1-nano' | 'openai/gpt-4o' | 'openai/gpt-4o-mini' | 'openai/gpt-4o-mini-search-preview' | 'openai/gpt-5' | 'openai/gpt-5-chat' | 'openai/gpt-5-codex' | 'openai/gpt-5-mini' | 'openai/gpt-5-nano' | 'openai/gpt-5-pro' | 'openai/gpt-5.1-codex' | 'openai/gpt-5.1-codex-max' | 'openai/gpt-5.1-codex-mini' | 'openai/gpt-5.1-instant' | 'openai/gpt-5.1-thinking' | 'openai/gpt-5.2' | 'openai/gpt-5.2-chat' | 'openai/gpt-5.2-codex' | 'openai/gpt-5.2-pro' | 'openai/gpt-5.3-chat' | 'openai/gpt-5.3-codex' | 'openai/gpt-5.4' | 'openai/gpt-5.4-mini' | 'openai/gpt-5.4-nano' | 'openai/gpt-5.4-pro' | 'openai/gpt-5.5' | 'openai/gpt-5.5-pro' | 'openai/gpt-oss-120b' | 'openai/gpt-oss-20b' | 'openai/gpt-oss-safeguard-20b' | 'openai/o1' | 'openai/o3' | 'openai/o3-deep-research' | 'openai/o3-mini' | 'openai/o3-pro' | 'openai/o4-mini' | 'perplexity/sonar' | 'perplexity/sonar-pro' | 'perplexity/sonar-reasoning-pro' | 'stepfun/step-3.5-flash' | 'stepfun/step-3.7-flash' | 'xai/grok-4.1-fast-non-reasoning' | 'xai/grok-4.1-fast-reasoning' | 'xai/grok-4.20-multi-agent' | 'xai/grok-4.20-multi-agent-beta' | 'xai/grok-4.20-non-reasoning' | 'xai/grok-4.20-non-reasoning-beta' | 'xai/grok-4.20-reasoning' | 'xai/grok-4.20-reasoning-beta' | 'xai/grok-4.3' | 'xai/grok-build-0.1' | 'xiaomi/mimo-v2-flash' | 'xiaomi/mimo-v2-pro' | 'xiaomi/mimo-v2.5' | 'xiaomi/mimo-v2.5-pro' | 'zai/glm-4.5' | 'zai/glm-4.5-air' | 'zai/glm-4.5v' | 'zai/glm-4.6' | 'zai/glm-4.6v' | 'zai/glm-4.6v-flash' | 'zai/glm-4.7' | 'zai/glm-4.7-flash' | 'zai/glm-4.7-flashx' | 'zai/glm-5' | 'zai/glm-5-turbo' | 'zai/glm-5.1' | 'zai/glm-5v-turbo' | (string & {});
 
-type GatewayVideoModelId = 'alibaba/wan-v2.5-t2v-preview' | 'alibaba/wan-v2.6-i2v' | 'alibaba/wan-v2.6-i2v-flash' | 'alibaba/wan-v2.6-r2v' | 'alibaba/wan-v2.6-r2v-flash' | 'alibaba/wan-v2.6-t2v' | 'bytedance/seedance-v1.0-lite-i2v' | 'bytedance/seedance-v1.0-lite-t2v' | 'bytedance/seedance-v1.0-pro' | 'bytedance/seedance-v1.0-pro-fast' | 'bytedance/seedance-v1.5-pro' | 'google/veo-3.0-fast-generate-001' | 'google/veo-3.0-generate-001' | 'google/veo-3.1-fast-generate-001' | 'google/veo-3.1-generate-001' | 'klingai/kling-v2.5-turbo-i2v' | 'klingai/kling-v2.5-turbo-t2v' | 'klingai/kling-v2.6-i2v' | 'klingai/kling-v2.6-motion-control' | 'klingai/kling-v2.6-t2v' | 'klingai/kling-v3.0-i2v' | 'klingai/kling-v3.0-t2v' | 'xai/grok-imagine-video' | (string & {});
+/**
+ * Shared WebSocket subprotocol contract for AI Gateway realtime auth.
+ *
+ * The browser `WebSocket` API cannot set request headers, so the Gateway auth
+ * (bearer) token is carried through the `Sec-WebSocket-Protocol` handshake
+ * instead of an `Authorization` header — the same workaround OpenAI uses for
+ * `openai-insecure-api-key.<token>`.
+ *
+ * This module is the single source of truth for that contract so the client and
+ * the Gateway server can't drift: the client encodes values with
+ * `getGatewayRealtimeProtocols`, and the Gateway server decodes them with
+ * `getGatewayRealtimeAuthToken` / `getGatewayRealtimeTeamIdOrSlug`.
+ *
+ * WebSocket subprotocol values must fit the RFC token grammar. The auth token is
+ * sent as-is, so callers must use tokens that are valid subprotocol tokens; the
+ * optional team scope is base64url-encoded by this module. Keep the complete
+ * `Sec-WebSocket-Protocol` header compact (target under an 8 KiB safe header
+ * budget) because intermediaries may reject large upgrade headers.
+ */
+/**
+ * Marker subprotocol offered on every handshake so the Gateway can echo a
+ * negotiated subprotocol on the 101 response (some clients require the server to
+ * select one of the offered subprotocols).
+ */
+declare const GATEWAY_REALTIME_SUBPROTOCOL = "ai-gateway-realtime.v1";
+/** Subprotocol prefix that carries the Gateway auth (bearer) token. */
+declare const GATEWAY_AUTH_SUBPROTOCOL_PREFIX = "ai-gateway-auth.";
+/** Subprotocol prefix that carries optional Vercel team scoping. */
+declare const GATEWAY_TEAM_SUBPROTOCOL_PREFIX = "ai-gateway-team.";
+/**
+ * Client-side: build the WebSocket subprotocols that carry `token` to the
+ * Gateway. Pass the result as the second argument to `new WebSocket(url, ...)`.
+ */
+declare function getGatewayRealtimeProtocols(token: string, options?: {
+    teamIdOrSlug?: string;
+}): string[];
+/**
+ * Server-side: extract the Gateway auth (bearer) token from a
+ * `Sec-WebSocket-Protocol` header value, or `undefined` when it is absent or
+ * empty. The Gateway upgrade handler turns this into an
+ * `Authorization: Bearer <token>` before its normal auth path.
+ *
+ * Accepts the raw header value (subprotocols are comma-separated and may carry
+ * surrounding whitespace).
+ */
+declare function getGatewayRealtimeAuthToken(secWebSocketProtocol: string | null | undefined): string | undefined;
+/**
+ * Server-side: extract the optional Vercel team ID or slug from the
+ * `Sec-WebSocket-Protocol` header value. Team scoping is base64url-encoded so
+ * arbitrary team slugs stay within the WebSocket subprotocol token grammar.
+ */
+declare function getGatewayRealtimeTeamIdOrSlug(secWebSocketProtocol: string | null | undefined): string | undefined;
 
+type GatewayRealtimeModelId = string & {};
+
+type GatewayRerankingModelId = 'cohere/rerank-v3.5' | 'cohere/rerank-v4-fast' | 'cohere/rerank-v4-pro' | 'voyage/rerank-2.5' | 'voyage/rerank-2.5-lite' | (string & {});
+
+type GatewaySpeechModelId = string & {};
+
+type GatewayTranscriptionModelId = string & {};
+
+type GatewayVideoModelId = 'alibaba/wan-v2.5-t2v-preview' | 'alibaba/wan-v2.6-i2v' | 'alibaba/wan-v2.6-i2v-flash' | 'alibaba/wan-v2.6-r2v' | 'alibaba/wan-v2.6-r2v-flash' | 'alibaba/wan-v2.6-t2v' | 'bytedance/seedance-2.0' | 'bytedance/seedance-2.0-fast' | 'bytedance/seedance-v1.0-lite-i2v' | 'bytedance/seedance-v1.0-lite-t2v' | 'bytedance/seedance-v1.0-pro' | 'bytedance/seedance-v1.0-pro-fast' | 'bytedance/seedance-v1.5-pro' | 'google/veo-3.0-fast-generate-001' | 'google/veo-3.0-generate-001' | 'google/veo-3.1-fast-generate-001' | 'google/veo-3.1-generate-001' | 'klingai/kling-v2.5-turbo-i2v' | 'klingai/kling-v2.5-turbo-t2v' | 'klingai/kling-v2.6-i2v' | 'klingai/kling-v2.6-motion-control' | 'klingai/kling-v2.6-t2v' | 'klingai/kling-v3.0-i2v' | 'klingai/kling-v3.0-motion-control' | 'klingai/kling-v3.0-t2v' | 'xai/grok-imagine-video' | 'xai/grok-imagine-video-1.5-preview' | (string & {});
+
+declare const KNOWN_MODEL_TYPES: readonly ["embedding", "image", "language", "reranking", "speech", "transcription", "video"];
+type KnownModelType = (typeof KNOWN_MODEL_TYPES)[number];
 interface GatewayLanguageModelEntry {
     /**
      * The model id used by the remote provider in model settings and for specifying the
@@ -50,9 +113,9 @@ interface GatewayLanguageModelEntry {
     /**
      * Optional field to differentiate between model types.
      */
-    modelType?: 'language' | 'embedding' | 'image' | 'video' | null;
+    modelType?: KnownModelType | null;
 }
-type GatewayLanguageModelSpecification = Pick<LanguageModelV3, 'specificationVersion' | 'provider' | 'modelId'>;
+type GatewayLanguageModelSpecification = Pick<LanguageModelV4, 'specificationVersion' | 'provider' | 'modelId'>;
 
 interface GatewayFetchMetadataResponse {
     models: GatewayLanguageModelEntry[];
@@ -64,9 +127,108 @@ interface GatewayCreditsResponse {
     totalUsed: string;
 }
 
-type GatewayEmbeddingModelId = 'alibaba/qwen3-embedding-0.6b' | 'alibaba/qwen3-embedding-4b' | 'alibaba/qwen3-embedding-8b' | 'amazon/titan-embed-text-v2' | 'cohere/embed-v4.0' | 'google/gemini-embedding-001' | 'google/text-embedding-005' | 'google/text-multilingual-embedding-002' | 'mistral/codestral-embed' | 'mistral/mistral-embed' | 'openai/text-embedding-3-large' | 'openai/text-embedding-3-small' | 'openai/text-embedding-ada-002' | 'voyage/voyage-3-large' | 'voyage/voyage-3.5' | 'voyage/voyage-3.5-lite' | 'voyage/voyage-4' | 'voyage/voyage-4-large' | 'voyage/voyage-4-lite' | 'voyage/voyage-code-2' | 'voyage/voyage-code-3' | 'voyage/voyage-finance-2' | 'voyage/voyage-law-2' | (string & {});
+interface GatewaySpendReportParams {
+    /** Start date in YYYY-MM-DD format (inclusive) */
+    startDate: string;
+    /** End date in YYYY-MM-DD format (inclusive) */
+    endDate: string;
+    /** Primary aggregation dimension. Defaults to 'day'. */
+    groupBy?: 'day' | 'user' | 'model' | 'tag' | 'provider' | 'credential_type';
+    /** Time granularity when groupBy is 'day'. */
+    datePart?: 'day' | 'hour';
+    /** Filter to a specific user's spend. */
+    userId?: string;
+    /** Filter to a specific model (e.g. 'anthropic/claude-sonnet-4.5'). */
+    model?: string;
+    /** Filter to a specific provider (e.g. 'anthropic'). */
+    provider?: string;
+    /** Filter to BYOK or system credentials. */
+    credentialType?: 'byok' | 'system';
+    /** Filter to requests with these tags. */
+    tags?: string[];
+}
+interface GatewaySpendReportRow {
+    /** Date string (present when groupBy is 'day') */
+    day?: string;
+    /** Hour timestamp (present when groupBy is 'day' and datePart is 'hour') */
+    hour?: string;
+    /** User identifier (present when groupBy is 'user') */
+    user?: string;
+    /** Model identifier (present when groupBy is 'model') */
+    model?: string;
+    /** Tag value (present when groupBy is 'tag') */
+    tag?: string;
+    /** Provider name (present when groupBy is 'provider') */
+    provider?: string;
+    /** Credential type (present when groupBy is 'credential_type') */
+    credentialType?: 'byok' | 'system';
+    /** Total cost in USD */
+    totalCost: number;
+    /** Market cost in USD */
+    marketCost?: number;
+    /** Number of input tokens */
+    inputTokens?: number;
+    /** Number of output tokens */
+    outputTokens?: number;
+    /** Number of cached input tokens */
+    cachedInputTokens?: number;
+    /** Number of cache creation input tokens */
+    cacheCreationInputTokens?: number;
+    /** Number of reasoning tokens */
+    reasoningTokens?: number;
+    /** Number of requests */
+    requestCount?: number;
+}
+interface GatewaySpendReportResponse {
+    results: GatewaySpendReportRow[];
+}
+
+interface GatewayGenerationInfoParams {
+    /** The generation ID to look up (format: gen_<ulid>) */
+    id: string;
+}
+interface GatewayGenerationInfo {
+    /** The generation ID */
+    id: string;
+    /** Total cost in USD */
+    totalCost: number;
+    /** Upstream inference cost in USD (BYOK only) */
+    upstreamInferenceCost: number;
+    /** Usage cost in USD (same as totalCost) */
+    usage: number;
+    /** ISO 8601 timestamp when the generation was created */
+    createdAt: string;
+    /** Model identifier */
+    model: string;
+    /** Whether BYOK credentials were used */
+    isByok: boolean;
+    /** Provider that served this generation */
+    providerName: string;
+    /** Whether streaming was used */
+    streamed: boolean;
+    /** Finish reason (e.g. 'stop') */
+    finishReason: string;
+    /** Time to first token in milliseconds */
+    latency: number;
+    /** Total generation time in milliseconds */
+    generationTime: number;
+    /** Number of prompt tokens */
+    promptTokens: number;
+    /** Number of completion tokens */
+    completionTokens: number;
+    /** Reasoning tokens used */
+    reasoningTokens: number;
+    /** Cached tokens used */
+    cachedTokens: number;
+    /** Cache creation input tokens */
+    cacheCreationTokens: number;
+    /** Billable web search calls */
+    billableWebSearchCalls: number;
+}
+
+type GatewayEmbeddingModelId = 'alibaba/qwen3-embedding-0.6b' | 'alibaba/qwen3-embedding-4b' | 'alibaba/qwen3-embedding-8b' | 'amazon/titan-embed-text-v2' | 'cohere/embed-v4.0' | 'google/gemini-embedding-001' | 'google/gemini-embedding-2' | 'google/text-embedding-005' | 'google/text-multilingual-embedding-002' | 'mistral/codestral-embed' | 'mistral/mistral-embed' | 'openai/text-embedding-3-large' | 'openai/text-embedding-3-small' | 'openai/text-embedding-ada-002' | 'voyage/voyage-3-large' | 'voyage/voyage-3.5' | 'voyage/voyage-3.5-lite' | 'voyage/voyage-4' | 'voyage/voyage-4-large' | 'voyage/voyage-4-lite' | 'voyage/voyage-code-2' | 'voyage/voyage-code-3' | 'voyage/voyage-finance-2' | 'voyage/voyage-law-2' | (string & {});
 
-type GatewayImageModelId = 'bfl/flux-kontext-max' | 'bfl/flux-kontext-pro' | 'bfl/flux-pro-1.0-fill' | 'bfl/flux-pro-1.1' | 'bfl/flux-pro-1.1-ultra' | 'google/imagen-4.0-fast-generate-001' | 'google/imagen-4.0-generate-001' | 'google/imagen-4.0-ultra-generate-001' | 'openai/gpt-image-1' | 'openai/gpt-image-1-mini' | 'openai/gpt-image-1.5' | 'recraft/recraft-v2' | 'recraft/recraft-v3' | 'recraft/recraft-v4' | 'recraft/recraft-v4-pro' | 'xai/grok-imagine-image' | 'xai/grok-imagine-image-pro' | (string & {});
+type GatewayImageModelId = 'bfl/flux-2-flex' | 'bfl/flux-2-klein-4b' | 'bfl/flux-2-klein-9b' | 'bfl/flux-2-max' | 'bfl/flux-2-pro' | 'bfl/flux-kontext-max' | 'bfl/flux-kontext-pro' | 'bfl/flux-pro-1.0-fill' | 'bfl/flux-pro-1.1' | 'bfl/flux-pro-1.1-ultra' | 'bytedance/seedream-4.0' | 'bytedance/seedream-4.5' | 'bytedance/seedream-5.0-lite' | 'google/imagen-4.0-fast-generate-001' | 'google/imagen-4.0-generate-001' | 'google/imagen-4.0-ultra-generate-001' | 'openai/gpt-image-1' | 'openai/gpt-image-1-mini' | 'openai/gpt-image-1.5' | 'openai/gpt-image-2' | 'prodia/flux-fast-schnell' | 'recraft/recraft-v2' | 'recraft/recraft-v3' | 'recraft/recraft-v4' | 'recraft/recraft-v4-pro' | 'recraft/recraft-v4.1' | 'recraft/recraft-v4.1-pro' | 'recraft/recraft-v4.1-utility' | 'recraft/recraft-v4.1-utility-pro' | 'xai/grok-imagine-image' | (string & {});
 
 interface PerplexitySearchConfig {
     /**
@@ -194,7 +356,7 @@ interface PerplexitySearchInput {
     search_recency_filter?: 'day' | 'week' | 'month' | 'year';
 }
 type PerplexitySearchOutput = PerplexitySearchResponse | PerplexitySearchError;
-declare const perplexitySearchToolFactory: _ai_sdk_provider_utils.ProviderToolFactoryWithOutputSchema<PerplexitySearchInput, PerplexitySearchOutput, PerplexitySearchConfig>;
+declare const perplexitySearchToolFactory: _ai_sdk_provider_utils.ProviderExecutedToolFactory<PerplexitySearchInput, PerplexitySearchOutput, PerplexitySearchConfig, {}>;
 
 interface ParallelSearchSourcePolicy {
     /**
@@ -327,7 +489,7 @@ interface ParallelSearchInput {
     };
 }
 type ParallelSearchOutput = ParallelSearchResponse | ParallelSearchError;
-declare const parallelSearchToolFactory: _ai_sdk_provider_utils.ProviderToolFactoryWithOutputSchema<ParallelSearchInput, ParallelSearchOutput, ParallelSearchConfig>;
+declare const parallelSearchToolFactory: _ai_sdk_provider_utils.ProviderExecutedToolFactory<ParallelSearchInput, ParallelSearchOutput, ParallelSearchConfig, {}>;
 
 /**
  * Gateway-specific provider-defined tools.
@@ -352,16 +514,16 @@ declare const gatewayTools: {
     perplexitySearch: (config?: PerplexitySearchConfig) => ReturnType<typeof perplexitySearchToolFactory>;
 };
 
-interface GatewayProvider extends ProviderV3 {
-    (modelId: GatewayModelId): LanguageModelV3;
+interface GatewayProvider extends ProviderV4 {
+    (modelId: GatewayModelId): LanguageModelV4;
     /**
      * Creates a model for text generation.
      */
-    chat(modelId: GatewayModelId): LanguageModelV3;
+    chat(modelId: GatewayModelId): LanguageModelV4;
     /**
      * Creates a model for text generation.
      */
-    languageModel(modelId: GatewayModelId): LanguageModelV3;
+    languageModel(modelId: GatewayModelId): LanguageModelV4;
     /**
      * Returns available providers and models for use with the remote provider.
      */
@@ -370,34 +532,73 @@ interface GatewayProvider extends ProviderV3 {
      * Returns credit information for the authenticated user.
      */
     getCredits(): Promise<GatewayCreditsResponse>;
+    /**
+     * Returns a spend report with cost, token, and request count data,
+     * aggregated by the specified dimension.
+     */
+    getSpendReport(params: GatewaySpendReportParams): Promise<GatewaySpendReportResponse>;
+    /**
+     * Returns detailed information about a specific generation by its ID,
+     * including cost, token usage, latency, and provider details.
+     */
+    getGenerationInfo(params: GatewayGenerationInfoParams): Promise<GatewayGenerationInfo>;
     /**
      * Creates a model for generating text embeddings.
      */
-    embedding(modelId: GatewayEmbeddingModelId): EmbeddingModelV3;
+    embedding(modelId: GatewayEmbeddingModelId): EmbeddingModelV4;
     /**
      * Creates a model for generating text embeddings.
      */
-    embeddingModel(modelId: GatewayEmbeddingModelId): EmbeddingModelV3;
+    embeddingModel(modelId: GatewayEmbeddingModelId): EmbeddingModelV4;
     /**
      * @deprecated Use `embeddingModel` instead.
      */
-    textEmbeddingModel(modelId: GatewayEmbeddingModelId): EmbeddingModelV3;
+    textEmbeddingModel(modelId: GatewayEmbeddingModelId): EmbeddingModelV4;
     /**
      * Creates a model for generating images.
      */
-    image(modelId: GatewayImageModelId): ImageModelV3;
+    image(modelId: GatewayImageModelId): ImageModelV4;
     /**
      * Creates a model for generating images.
      */
-    imageModel(modelId: GatewayImageModelId): ImageModelV3;
+    imageModel(modelId: GatewayImageModelId): ImageModelV4;
     /**
      * Creates a model for generating videos.
      */
-    video(modelId: GatewayVideoModelId): Experimental_VideoModelV3;
+    video(modelId: GatewayVideoModelId): Experimental_VideoModelV4;
     /**
      * Creates a model for generating videos.
      */
-    videoModel(modelId: GatewayVideoModelId): Experimental_VideoModelV3;
+    videoModel(modelId: GatewayVideoModelId): Experimental_VideoModelV4;
+    /**
+     * Creates a model for reranking documents.
+     */
+    reranking(modelId: GatewayRerankingModelId): RerankingModelV4;
+    /**
+     * Creates a model for reranking documents.
+     */
+    rerankingModel(modelId: GatewayRerankingModelId): RerankingModelV4;
+    /**
+     * Creates a model for text-to-speech generation.
+     */
+    speech(modelId: GatewaySpeechModelId): SpeechModelV4;
+    /**
+     * Creates a model for text-to-speech generation.
+     */
+    speechModel(modelId: GatewaySpeechModelId): SpeechModelV4;
+    /**
+     * Creates a model for audio transcription.
+     */
+    transcription(modelId: GatewayTranscriptionModelId): TranscriptionModelV4;
+    /**
+     * Creates a model for audio transcription.
+     */
+    transcriptionModel(modelId: GatewayTranscriptionModelId): TranscriptionModelV4;
+    /**
+     * Creates an experimental realtime model for bidirectional audio/text
+     * communication over WebSocket, normalized through the AI Gateway.
+     */
+    experimental_realtime: Experimental_RealtimeFactoryV4;
     /**
      * Gateway-specific tools executed server-side.
      */
@@ -405,13 +606,19 @@ interface GatewayProvider extends ProviderV3 {
 }
 interface GatewayProviderSettings {
     /**
-     * The base URL prefix for API calls. Defaults to `https://ai-gateway.vercel.sh/v1/ai`.
+     * The base URL prefix for API calls. Defaults to `https://ai-gateway.vercel.sh/v4/ai`.
      */
     baseURL?: string;
     /**
-     * API key that is being sent using the `Authorization` header.
+     * API key or Vercel access token that is being sent using the `Authorization`
+     * header. It defaults to the `AI_GATEWAY_API_KEY` environment variable.
      */
     apiKey?: string;
+    /**
+     * Vercel team ID or slug to scope requests for access tokens that can access
+     * multiple teams.
+     */
+    teamIdOrSlug?: string;
     /**
      * Custom headers to include in the requests.
      */
@@ -429,36 +636,62 @@ interface GatewayProviderSettings {
 /**
  * Create a remote provider instance.
  */
-declare function createGatewayProvider(options?: GatewayProviderSettings): GatewayProvider;
+declare function createGateway(options?: GatewayProviderSettings): GatewayProvider;
 declare const gateway: GatewayProvider;
 
-declare const gatewayLanguageModelOptions: _ai_sdk_provider_utils.LazySchema<{
-    only?: string[] | undefined;
-    order?: string[] | undefined;
-    user?: string | undefined;
-    tags?: string[] | undefined;
-    models?: string[] | undefined;
-    byok?: Record<string, Record<string, unknown>[]> | undefined;
-    zeroDataRetention?: boolean | undefined;
+type GatewayProviderOptions = {
+    /**
+     * Service-owned options may be added by the Gateway without requiring an SDK
+     * release. The Gateway service validates and applies the runtime schema.
+     */
+    [key: string]: unknown;
+    /** Request-scoped BYOK credentials to use instead of cached credentials. */
+    byok?: Record<string, Array<Record<string, unknown>>>;
+    /** Enables automatic caching behavior when supported by the Gateway. */
+    caching?: 'auto';
+    /** Filter to providers that do not train on prompt data. */
+    disallowPromptTraining?: boolean;
+    /** Filter to providers that are HIPAA compliant with Vercel AI Gateway. */
+    hipaaCompliant?: boolean;
+    /** Array of model slugs specifying fallback models to use in order. */
+    models?: string[];
+    /** Array of provider slugs that are the only ones allowed to be used. */
+    only?: string[];
+    /** Array of provider slugs specifying the provider attempt order. */
+    order?: string[];
+    /** Per-provider timeouts for BYOK credentials in milliseconds. */
     providerTimeouts?: {
-        byok?: Record<string, number> | undefined;
-    } | undefined;
-}>;
-type GatewayLanguageModelOptions = InferSchema<typeof gatewayLanguageModelOptions>;
+        byok?: Record<string, number>;
+    };
+    /** Entity identifier against which quota is tracked. */
+    quotaEntityId?: string;
+    /** Unified service tier intent. */
+    serviceTier?: 'flex' | 'priority';
+    /** Sort providers by a performance or cost metric before routing. */
+    sort?: 'cost' | 'tps' | 'ttft';
+    /** User-specified tags for reporting and filtering usage. */
+    tags?: string[];
+    /** End-user identifier for spend tracking and attribution. */
+    user?: string;
+    /** Filter to providers with zero data retention agreements. */
+    zeroDataRetention?: boolean;
+};
 
-declare const symbol$6: unique symbol;
+declare const symbol$8: unique symbol;
 declare abstract class GatewayError extends Error {
-    private readonly [symbol$6];
+    private readonly [symbol$8];
     abstract readonly name: string;
     abstract readonly type: string;
     readonly statusCode: number;
     readonly cause?: unknown;
     readonly generationId?: string;
-    constructor({ message, statusCode, cause, generationId, }: {
+    readonly isRetryable: boolean;
+    constructor({ message, statusCode, cause, generationId, isRetryable, }: {
         message: string;
         statusCode?: number;
         cause?: unknown;
         generationId?: string;
+        isRetryable?: boolean;
     });
     /**
      * Checks if the given error is a Gateway Error.
@@ -480,12 +713,12 @@ declare const gatewayErrorResponseSchema: _ai_sdk_provider_utils.LazySchema<{
 }>;
 type GatewayErrorResponse = InferSchema<typeof gatewayErrorResponseSchema>;
 
-declare const symbol$5: unique symbol;
+declare const symbol$7: unique symbol;
 /**
  * Authentication failed - invalid API key or OIDC token
  */
 declare class GatewayAuthenticationError extends GatewayError {
-    private readonly [symbol$5];
+    private readonly [symbol$7];
     readonly name = "GatewayAuthenticationError";
     readonly type = "authentication_error";
     constructor({ message, statusCode, cause, generationId, }?: {
@@ -498,7 +731,7 @@ declare class GatewayAuthenticationError extends GatewayError {
     /**
      * Creates a contextual error message when authentication fails
      */
-    static createContextualError({ apiKeyProvided, oidcTokenProvided, message, statusCode, cause, generationId, }: {
+    static createContextualError({ apiKeyProvided, oidcTokenProvided, statusCode, cause, generationId, }: {
         apiKeyProvided: boolean;
         oidcTokenProvided: boolean;
         message?: string;
@@ -508,6 +741,43 @@ declare class GatewayAuthenticationError extends GatewayError {
     }): GatewayAuthenticationError;
 }
 
+declare const symbol$6: unique symbol;
+/**
+ * The request could not be fulfilled because a dependency it relied on was not
+ * available on the credentials or provider used to serve it (HTTP 424). Not
+ * retryable — the caller must change the request.
+ */
+declare class GatewayFailedDependencyError extends GatewayError {
+    private readonly [symbol$6];
+    readonly name = "GatewayFailedDependencyError";
+    readonly type = "failed_dependency";
+    constructor({ message, statusCode, cause, generationId, }?: {
+        message?: string;
+        statusCode?: number;
+        cause?: unknown;
+        generationId?: string;
+    });
+    static isInstance(error: unknown): error is GatewayFailedDependencyError;
+}
+
+declare const symbol$5: unique symbol;
+/**
+ * Forbidden - the request was rejected by policy (e.g. a routing rule),
+ * not an authentication failure.
+ */
+declare class GatewayForbiddenError extends GatewayError {
+    private readonly [symbol$5];
+    readonly name = "GatewayForbiddenError";
+    readonly type = "forbidden";
+    constructor({ message, statusCode, cause, generationId, }?: {
+        message?: string;
+        statusCode?: number;
+        cause?: unknown;
+        generationId?: string;
+    });
+    static isInstance(error: unknown): error is GatewayForbiddenError;
+}
+
 declare const symbol$4: unique symbol;
 /**
  * Internal server error from the Gateway
@@ -599,4 +869,6 @@ declare class GatewayResponseError extends GatewayError {
     static isInstance(error: unknown): error is GatewayResponseError;
 }
 
-export { GatewayAuthenticationError, type GatewayCreditsResponse, GatewayError, type GatewayErrorResponse, GatewayInternalServerError, GatewayInvalidRequestError, type GatewayLanguageModelEntry, type GatewayLanguageModelOptions, type GatewayLanguageModelSpecification, type GatewayLanguageModelEntry as GatewayModelEntry, type GatewayModelId, GatewayModelNotFoundError, type GatewayProvider, type GatewayLanguageModelOptions as GatewayProviderOptions, type GatewayProviderSettings, GatewayRateLimitError, GatewayResponseError, type GatewayVideoModelId, createGatewayProvider as createGateway, createGatewayProvider, gateway };
+declare const VERSION: string;
+
+export { GATEWAY_AUTH_SUBPROTOCOL_PREFIX, GATEWAY_REALTIME_SUBPROTOCOL, GATEWAY_TEAM_SUBPROTOCOL_PREFIX, GatewayAuthenticationError, type GatewayCreditsResponse, GatewayError, type GatewayErrorResponse, GatewayFailedDependencyError, GatewayForbiddenError, type GatewayGenerationInfo, type GatewayGenerationInfoParams, GatewayInternalServerError, GatewayInvalidRequestError, type GatewayLanguageModelEntry, type GatewayProviderOptions as GatewayLanguageModelOptions, type GatewayLanguageModelSpecification, type GatewayLanguageModelEntry as GatewayModelEntry, type GatewayModelId, GatewayModelNotFoundError, type GatewayProvider, type GatewayProviderOptions, type GatewayProviderSettings, GatewayRateLimitError, type GatewayRealtimeModelId, type GatewayRerankingModelId, GatewayResponseError, type GatewaySpeechModelId, type GatewaySpendReportParams, type GatewaySpendReportResponse, type GatewaySpendReportRow, type GatewayTranscriptionModelId, type GatewayVideoModelId, VERSION, createGateway, createGateway as createGatewayProvider, gateway, getGatewayRealtimeAuthToken, getGatewayRealtimeProtocols, getGatewayRealtimeTeamIdOrSlug };