@ai-sdk/gateway 4.0.0-canary.86 → 4.0.0-canary.88

package/CHANGELOG.md

@@ -1,5 +1,17 @@
 # @ai-sdk/gateway
 
+## 4.0.0-canary.88
+
+### Patch Changes
+
+- accaca0: chore(provider/gateway): update gateway model settings files
+
+## 4.0.0-canary.87
+
+### Patch Changes
+
+- bf837fe: feat(provider/gateway): add speech and transcription model support
+
 ## 4.0.0-canary.86
 
 ### Patch Changes

package/dist/index.d.ts

@@ -1,14 +1,18 @@
-import { LanguageModelV4, ProviderV4, EmbeddingModelV4, ImageModelV4, Experimental_VideoModelV4, RerankingModelV4, TypeValidationError } from '@ai-sdk/provider';
+import { LanguageModelV4, ProviderV4, EmbeddingModelV4, ImageModelV4, Experimental_VideoModelV4, RerankingModelV4, SpeechModelV4, TranscriptionModelV4, 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/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' | '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-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-2.0-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-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' | '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' | '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/pixtral-12b' | 'mistral/pixtral-large' | 'moonshotai/kimi-k2' | 'moonshotai/kimi-k2-thinking' | 'moonshotai/kimi-k2-thinking-turbo' | 'moonshotai/kimi-k2-turbo' | 'moonshotai/kimi-k2.5' | 'moonshotai/kimi-k2.6' | 'morph/morph-v3-fast' | 'morph/morph-v3-large' | 'nvidia/nemotron-3-nano-30b-a3b' | 'nvidia/nemotron-3-super-120b-a12b' | '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' | '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' | '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 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' | '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-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-2.0-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-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' | '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' | '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-thinking-turbo' | 'moonshotai/kimi-k2-turbo' | 'moonshotai/kimi-k2.5' | 'moonshotai/kimi-k2.6' | 'morph/morph-v3-fast' | 'morph/morph-v3-large' | 'nvidia/nemotron-3-nano-30b-a3b' | 'nvidia/nemotron-3-super-120b-a12b' | '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' | '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' | '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 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' | (string & {});
 
-declare const KNOWN_MODEL_TYPES: readonly ["embedding", "image", "language", "reranking", "video"];
+declare const KNOWN_MODEL_TYPES: readonly ["embedding", "image", "language", "reranking", "speech", "transcription", "video"];
 type KnownModelType = (typeof KNOWN_MODEL_TYPES)[number];
 interface GatewayLanguageModelEntry {
     /**
@@ -519,6 +523,22 @@ interface GatewayProvider extends ProviderV4 {
      * 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;
     /**
      * Gateway-specific tools executed server-side.
      */
@@ -728,4 +748,4 @@ declare class GatewayResponseError extends GatewayError {
 
 declare const VERSION: string;
 
-export { GatewayAuthenticationError, type GatewayCreditsResponse, GatewayError, type GatewayErrorResponse, 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 GatewayRerankingModelId, GatewayResponseError, type GatewaySpendReportParams, type GatewaySpendReportResponse, type GatewaySpendReportRow, type GatewayVideoModelId, VERSION, createGateway, createGateway as createGatewayProvider, gateway };
+export { GatewayAuthenticationError, type GatewayCreditsResponse, GatewayError, type GatewayErrorResponse, 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 GatewayRerankingModelId, GatewayResponseError, type GatewaySpeechModelId, type GatewaySpendReportParams, type GatewaySpendReportResponse, type GatewaySpendReportRow, type GatewayTranscriptionModelId, type GatewayVideoModelId, VERSION, createGateway, createGateway as createGatewayProvider, gateway };

package/dist/index.js

@@ -483,6 +483,8 @@ var KNOWN_MODEL_TYPES = [
   "image",
   "language",
   "reranking",
+  "speech",
+  "transcription",
   "video"
 ];
 
@@ -1520,43 +1522,276 @@ var gatewayRerankingResponseSchema = lazySchema8(
   )
 );
 
+// src/gateway-speech-model.ts
+import {
+  combineHeaders as combineHeaders6,
+  createJsonErrorResponseHandler as createJsonErrorResponseHandler9,
+  createJsonResponseHandler as createJsonResponseHandler8,
+  postJsonToApi as postJsonToApi6,
+  resolve as resolve9
+} from "@ai-sdk/provider-utils";
+import { z as z12 } from "zod/v4";
+var GatewaySpeechModel = class {
+  constructor(modelId, config) {
+    this.modelId = modelId;
+    this.config = config;
+    this.specificationVersion = "v4";
+  }
+  get provider() {
+    return this.config.provider;
+  }
+  async doGenerate({
+    text,
+    voice,
+    outputFormat,
+    instructions,
+    speed,
+    language,
+    providerOptions,
+    headers,
+    abortSignal
+  }) {
+    var _a9;
+    const resolvedHeaders = this.config.headers ? await resolve9(this.config.headers) : void 0;
+    try {
+      const {
+        responseHeaders,
+        value: responseBody,
+        rawValue
+      } = await postJsonToApi6({
+        url: this.getUrl(),
+        headers: combineHeaders6(
+          resolvedHeaders,
+          headers != null ? headers : {},
+          this.getModelConfigHeaders(),
+          await resolve9(this.config.o11yHeaders)
+        ),
+        body: {
+          text,
+          ...voice && { voice },
+          ...outputFormat && { outputFormat },
+          ...instructions && { instructions },
+          ...speed != null && { speed },
+          ...language && { language },
+          ...providerOptions && { providerOptions }
+        },
+        successfulResponseHandler: createJsonResponseHandler8(
+          gatewaySpeechResponseSchema
+        ),
+        failedResponseHandler: createJsonErrorResponseHandler9({
+          errorSchema: z12.any(),
+          errorToMessage: (data) => data
+        }),
+        ...abortSignal && { abortSignal },
+        fetch: this.config.fetch
+      });
+      return {
+        audio: responseBody.audio,
+        warnings: (_a9 = responseBody.warnings) != null ? _a9 : [],
+        providerMetadata: responseBody.providerMetadata,
+        response: {
+          timestamp: /* @__PURE__ */ new Date(),
+          modelId: this.modelId,
+          headers: responseHeaders,
+          body: rawValue
+        }
+      };
+    } catch (error) {
+      throw await asGatewayError(
+        error,
+        await parseAuthMethod(resolvedHeaders != null ? resolvedHeaders : {})
+      );
+    }
+  }
+  getUrl() {
+    return `${this.config.baseURL}/speech-model`;
+  }
+  getModelConfigHeaders() {
+    return {
+      "ai-speech-model-specification-version": "4",
+      "ai-model-id": this.modelId
+    };
+  }
+};
+var providerMetadataEntrySchema3 = z12.object({}).catchall(z12.unknown());
+var gatewaySpeechWarningSchema = z12.discriminatedUnion("type", [
+  z12.object({
+    type: z12.literal("unsupported"),
+    feature: z12.string(),
+    details: z12.string().optional()
+  }),
+  z12.object({
+    type: z12.literal("compatibility"),
+    feature: z12.string(),
+    details: z12.string().optional()
+  }),
+  z12.object({
+    type: z12.literal("other"),
+    message: z12.string()
+  })
+]);
+var gatewaySpeechResponseSchema = z12.object({
+  audio: z12.string(),
+  warnings: z12.array(gatewaySpeechWarningSchema).optional(),
+  providerMetadata: z12.record(z12.string(), providerMetadataEntrySchema3).optional()
+});
+
+// src/gateway-transcription-model.ts
+import {
+  combineHeaders as combineHeaders7,
+  convertUint8ArrayToBase64 as convertUint8ArrayToBase643,
+  createJsonErrorResponseHandler as createJsonErrorResponseHandler10,
+  createJsonResponseHandler as createJsonResponseHandler9,
+  postJsonToApi as postJsonToApi7,
+  resolve as resolve10
+} from "@ai-sdk/provider-utils";
+import { z as z13 } from "zod/v4";
+var GatewayTranscriptionModel = class {
+  constructor(modelId, config) {
+    this.modelId = modelId;
+    this.config = config;
+    this.specificationVersion = "v4";
+  }
+  get provider() {
+    return this.config.provider;
+  }
+  async doGenerate({
+    audio,
+    mediaType,
+    providerOptions,
+    headers,
+    abortSignal
+  }) {
+    var _a9, _b9, _c, _d;
+    const resolvedHeaders = this.config.headers ? await resolve10(this.config.headers) : void 0;
+    try {
+      const {
+        responseHeaders,
+        value: responseBody,
+        rawValue
+      } = await postJsonToApi7({
+        url: this.getUrl(),
+        headers: combineHeaders7(
+          resolvedHeaders,
+          headers != null ? headers : {},
+          this.getModelConfigHeaders(),
+          await resolve10(this.config.o11yHeaders)
+        ),
+        body: {
+          audio: audio instanceof Uint8Array ? convertUint8ArrayToBase643(audio) : audio,
+          mediaType,
+          ...providerOptions && { providerOptions }
+        },
+        successfulResponseHandler: createJsonResponseHandler9(
+          gatewayTranscriptionResponseSchema
+        ),
+        failedResponseHandler: createJsonErrorResponseHandler10({
+          errorSchema: z13.any(),
+          errorToMessage: (data) => data
+        }),
+        ...abortSignal && { abortSignal },
+        fetch: this.config.fetch
+      });
+      return {
+        text: responseBody.text,
+        segments: (_a9 = responseBody.segments) != null ? _a9 : [],
+        language: (_b9 = responseBody.language) != null ? _b9 : void 0,
+        durationInSeconds: (_c = responseBody.durationInSeconds) != null ? _c : void 0,
+        warnings: (_d = responseBody.warnings) != null ? _d : [],
+        providerMetadata: responseBody.providerMetadata,
+        response: {
+          timestamp: /* @__PURE__ */ new Date(),
+          modelId: this.modelId,
+          headers: responseHeaders,
+          body: rawValue
+        }
+      };
+    } catch (error) {
+      throw await asGatewayError(
+        error,
+        await parseAuthMethod(resolvedHeaders != null ? resolvedHeaders : {})
+      );
+    }
+  }
+  getUrl() {
+    return `${this.config.baseURL}/transcription-model`;
+  }
+  getModelConfigHeaders() {
+    return {
+      "ai-transcription-model-specification-version": "4",
+      "ai-model-id": this.modelId
+    };
+  }
+};
+var providerMetadataEntrySchema4 = z13.object({}).catchall(z13.unknown());
+var gatewayTranscriptionWarningSchema = z13.discriminatedUnion("type", [
+  z13.object({
+    type: z13.literal("unsupported"),
+    feature: z13.string(),
+    details: z13.string().optional()
+  }),
+  z13.object({
+    type: z13.literal("compatibility"),
+    feature: z13.string(),
+    details: z13.string().optional()
+  }),
+  z13.object({
+    type: z13.literal("other"),
+    message: z13.string()
+  })
+]);
+var gatewayTranscriptionResponseSchema = z13.object({
+  text: z13.string(),
+  segments: z13.array(
+    z13.object({
+      text: z13.string(),
+      startSecond: z13.number(),
+      endSecond: z13.number()
+    })
+  ).optional(),
+  language: z13.string().nullish(),
+  durationInSeconds: z13.number().nullish(),
+  warnings: z13.array(gatewayTranscriptionWarningSchema).optional(),
+  providerMetadata: z13.record(z13.string(), providerMetadataEntrySchema4).optional()
+});
+
 // src/tool/parallel-search.ts
 import {
   createProviderExecutedToolFactory,
   lazySchema as lazySchema9,
   zodSchema as zodSchema9
 } from "@ai-sdk/provider-utils";
-import { z as z12 } from "zod";
+import { z as z14 } from "zod";
 var parallelSearchInputSchema = lazySchema9(
   () => zodSchema9(
-    z12.object({
-      objective: z12.string().describe(
+    z14.object({
+      objective: z14.string().describe(
         "Natural-language description of the web research goal, including source or freshness guidance and broader context from the task. Maximum 5000 characters."
       ),
-      search_queries: z12.array(z12.string()).optional().describe(
+      search_queries: z14.array(z14.string()).optional().describe(
         "Optional search queries to supplement the objective. Maximum 200 characters per query."
       ),
-      mode: z12.enum(["one-shot", "agentic"]).optional().describe(
+      mode: z14.enum(["one-shot", "agentic"]).optional().describe(
         'Mode preset: "one-shot" for comprehensive results with longer excerpts (default), "agentic" for concise, token-efficient results for multi-step workflows.'
       ),
-      max_results: z12.number().optional().describe(
+      max_results: z14.number().optional().describe(
         "Maximum number of results to return (1-20). Defaults to 10 if not specified."
       ),
-      source_policy: z12.object({
-        include_domains: z12.array(z12.string()).optional().describe("List of domains to include in search results."),
-        exclude_domains: z12.array(z12.string()).optional().describe("List of domains to exclude from search results."),
-        after_date: z12.string().optional().describe(
+      source_policy: z14.object({
+        include_domains: z14.array(z14.string()).optional().describe("List of domains to include in search results."),
+        exclude_domains: z14.array(z14.string()).optional().describe("List of domains to exclude from search results."),
+        after_date: z14.string().optional().describe(
           "Only include results published after this date (ISO 8601 format)."
         )
       }).optional().describe(
         "Source policy for controlling which domains to include/exclude and freshness."
       ),
-      excerpts: z12.object({
-        max_chars_per_result: z12.number().optional().describe("Maximum characters per result."),
-        max_chars_total: z12.number().optional().describe("Maximum total characters across all results.")
+      excerpts: z14.object({
+        max_chars_per_result: z14.number().optional().describe("Maximum characters per result."),
+        max_chars_total: z14.number().optional().describe("Maximum total characters across all results.")
       }).optional().describe("Excerpt configuration for controlling result length."),
-      fetch_policy: z12.object({
-        max_age_seconds: z12.number().optional().describe(
+      fetch_policy: z14.object({
+        max_age_seconds: z14.number().optional().describe(
           "Maximum age in seconds for cached content. Set to 0 to always fetch fresh content."
         )
       }).optional().describe("Fetch policy for controlling content freshness.")
@@ -1565,23 +1800,23 @@ var parallelSearchInputSchema = lazySchema9(
 );
 var parallelSearchOutputSchema = lazySchema9(
   () => zodSchema9(
-    z12.union([
+    z14.union([
       // Success response
-      z12.object({
-        searchId: z12.string(),
-        results: z12.array(
-          z12.object({
-            url: z12.string(),
-            title: z12.string(),
-            excerpt: z12.string(),
-            publishDate: z12.string().nullable().optional(),
-            relevanceScore: z12.number().optional()
+      z14.object({
+        searchId: z14.string(),
+        results: z14.array(
+          z14.object({
+            url: z14.string(),
+            title: z14.string(),
+            excerpt: z14.string(),
+            publishDate: z14.string().nullable().optional(),
+            relevanceScore: z14.number().optional()
           })
         )
       }),
       // Error response
-      z12.object({
-        error: z12.enum([
+      z14.object({
+        error: z14.enum([
           "api_error",
           "rate_limit",
           "timeout",
@@ -1589,8 +1824,8 @@ var parallelSearchOutputSchema = lazySchema9(
           "configuration_error",
           "unknown"
         ]),
-        statusCode: z12.number().optional(),
-        message: z12.string()
+        statusCode: z14.number().optional(),
+        message: z14.string()
       })
     ])
   )
@@ -1608,44 +1843,44 @@ import {
   lazySchema as lazySchema10,
   zodSchema as zodSchema10
 } from "@ai-sdk/provider-utils";
-import { z as z13 } from "zod";
+import { z as z15 } from "zod";
 var perplexitySearchInputSchema = lazySchema10(
   () => zodSchema10(
-    z13.object({
-      query: z13.union([z13.string(), z13.array(z13.string())]).describe(
+    z15.object({
+      query: z15.union([z15.string(), z15.array(z15.string())]).describe(
         "Search query (string) or multiple queries (array of up to 5 strings). Multi-query searches return combined results from all queries."
       ),
-      max_results: z13.number().optional().describe(
+      max_results: z15.number().optional().describe(
         "Maximum number of search results to return (1-20, default: 10)"
       ),
-      max_tokens_per_page: z13.number().optional().describe(
+      max_tokens_per_page: z15.number().optional().describe(
         "Maximum number of tokens to extract per search result page (256-2048, default: 2048)"
       ),
-      max_tokens: z13.number().optional().describe(
+      max_tokens: z15.number().optional().describe(
         "Maximum total tokens across all search results (default: 25000, max: 1000000)"
       ),
-      country: z13.string().optional().describe(
+      country: z15.string().optional().describe(
         "Two-letter ISO 3166-1 alpha-2 country code for regional search results (e.g., 'US', 'GB', 'FR')"
       ),
-      search_domain_filter: z13.array(z13.string()).optional().describe(
+      search_domain_filter: z15.array(z15.string()).optional().describe(
         "List of domains to include or exclude from search results (max 20). To include: ['nature.com', 'science.org']. To exclude: ['-example.com', '-spam.net']"
       ),
-      search_language_filter: z13.array(z13.string()).optional().describe(
+      search_language_filter: z15.array(z15.string()).optional().describe(
         "List of ISO 639-1 language codes to filter results (max 10, lowercase). Examples: ['en', 'fr', 'de']"
       ),
-      search_after_date: z13.string().optional().describe(
+      search_after_date: z15.string().optional().describe(
         "Include only results published after this date. Format: 'MM/DD/YYYY' (e.g., '3/1/2025'). Cannot be used with search_recency_filter."
       ),
-      search_before_date: z13.string().optional().describe(
+      search_before_date: z15.string().optional().describe(
         "Include only results published before this date. Format: 'MM/DD/YYYY' (e.g., '3/15/2025'). Cannot be used with search_recency_filter."
       ),
-      last_updated_after_filter: z13.string().optional().describe(
+      last_updated_after_filter: z15.string().optional().describe(
         "Include only results last updated after this date. Format: 'MM/DD/YYYY' (e.g., '3/1/2025'). Cannot be used with search_recency_filter."
       ),
-      last_updated_before_filter: z13.string().optional().describe(
+      last_updated_before_filter: z15.string().optional().describe(
         "Include only results last updated before this date. Format: 'MM/DD/YYYY' (e.g., '3/15/2025'). Cannot be used with search_recency_filter."
       ),
-      search_recency_filter: z13.enum(["day", "week", "month", "year"]).optional().describe(
+      search_recency_filter: z15.enum(["day", "week", "month", "year"]).optional().describe(
         "Filter results by relative time period. Cannot be used with search_after_date or search_before_date."
       )
     })
@@ -1653,31 +1888,31 @@ var perplexitySearchInputSchema = lazySchema10(
 );
 var perplexitySearchOutputSchema = lazySchema10(
   () => zodSchema10(
-    z13.union([
+    z15.union([
       // Success response
-      z13.object({
-        results: z13.array(
-          z13.object({
-            title: z13.string(),
-            url: z13.string(),
-            snippet: z13.string(),
-            date: z13.string().optional(),
-            lastUpdated: z13.string().optional()
+      z15.object({
+        results: z15.array(
+          z15.object({
+            title: z15.string(),
+            url: z15.string(),
+            snippet: z15.string(),
+            date: z15.string().optional(),
+            lastUpdated: z15.string().optional()
           })
         ),
-        id: z13.string()
+        id: z15.string()
       }),
       // Error response
-      z13.object({
-        error: z13.enum([
+      z15.object({
+        error: z15.enum([
           "api_error",
           "rate_limit",
           "timeout",
           "invalid_input",
           "unknown"
         ]),
-        statusCode: z13.number().optional(),
-        message: z13.string()
+        statusCode: z15.number().optional(),
+        message: z15.string()
       })
     ])
   )
@@ -1719,7 +1954,7 @@ async function getVercelRequestId() {
 }
 
 // src/version.ts
-var VERSION = true ? "4.0.0-canary.86" : "0.0.0-test";
+var VERSION = true ? "4.0.0-canary.88" : "0.0.0-test";
 
 // src/gateway-provider.ts
 var AI_GATEWAY_PROTOCOL_VERSION = "0.0.1";
@@ -1899,6 +2134,28 @@ function createGateway(options = {}) {
   };
   provider.rerankingModel = createRerankingModel;
   provider.reranking = createRerankingModel;
+  const createSpeechModel = (modelId) => {
+    return new GatewaySpeechModel(modelId, {
+      provider: "gateway",
+      baseURL,
+      headers: getHeaders,
+      fetch: options.fetch,
+      o11yHeaders: createO11yHeaders()
+    });
+  };
+  provider.speechModel = createSpeechModel;
+  provider.speech = createSpeechModel;
+  const createTranscriptionModel = (modelId) => {
+    return new GatewayTranscriptionModel(modelId, {
+      provider: "gateway",
+      baseURL,
+      headers: getHeaders,
+      fetch: options.fetch,
+      o11yHeaders: createO11yHeaders()
+    });
+  };
+  provider.transcriptionModel = createTranscriptionModel;
+  provider.transcription = createTranscriptionModel;
   provider.chat = provider.languageModel;
   provider.embedding = provider.embeddingModel;
   provider.image = provider.imageModel;