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

package/src/gateway-generation-info.ts

@@ -0,0 +1,149 @@
+import {
+  createJsonErrorResponseHandler,
+  createJsonResponseHandler,
+  getFromApi,
+  lazySchema,
+  resolve,
+  zodSchema,
+} from '@ai-sdk/provider-utils';
+import { z } from 'zod/v4';
+import { asGatewayError } from './errors';
+import type { GatewayConfig } from './gateway-config';
+
+export interface GatewayGenerationInfoParams {
+  /** The generation ID to look up (format: gen_<ulid>) */
+  id: string;
+}
+
+export 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;
+}
+
+export class GatewayGenerationInfoFetcher {
+  constructor(private readonly config: GatewayConfig) {}
+
+  async getGenerationInfo(
+    params: GatewayGenerationInfoParams,
+  ): Promise<GatewayGenerationInfo> {
+    try {
+      const baseUrl = new URL(this.config.baseURL);
+
+      const { value } = await getFromApi({
+        url: `${baseUrl.origin}/v1/generation?id=${encodeURIComponent(params.id)}`,
+        headers: this.config.headers
+          ? await resolve(this.config.headers)
+          : undefined,
+        successfulResponseHandler: createJsonResponseHandler(
+          gatewayGenerationInfoResponseSchema,
+        ),
+        failedResponseHandler: createJsonErrorResponseHandler({
+          errorSchema: z.any(),
+          errorToMessage: data => data,
+        }),
+        fetch: this.config.fetch,
+      });
+
+      return value;
+    } catch (error) {
+      throw await asGatewayError(error);
+    }
+  }
+}
+
+const gatewayGenerationInfoResponseSchema = lazySchema(() =>
+  zodSchema(
+    z
+      .object({
+        data: z
+          .object({
+            id: z.string(),
+            total_cost: z.number(),
+            upstream_inference_cost: z.number(),
+            usage: z.number(),
+            created_at: z.string(),
+            model: z.string(),
+            is_byok: z.boolean(),
+            provider_name: z.string(),
+            streamed: z.boolean(),
+            finish_reason: z.string(),
+            latency: z.number(),
+            generation_time: z.number(),
+            native_tokens_prompt: z.number(),
+            native_tokens_completion: z.number(),
+            native_tokens_reasoning: z.number(),
+            native_tokens_cached: z.number(),
+            native_tokens_cache_creation: z.number(),
+            billable_web_search_calls: z.number(),
+          })
+          .transform(
+            ({
+              total_cost,
+              upstream_inference_cost,
+              created_at,
+              is_byok,
+              provider_name,
+              finish_reason,
+              generation_time,
+              native_tokens_prompt,
+              native_tokens_completion,
+              native_tokens_reasoning,
+              native_tokens_cached,
+              native_tokens_cache_creation,
+              billable_web_search_calls,
+              ...rest
+            }) => ({
+              ...rest,
+              totalCost: total_cost,
+              upstreamInferenceCost: upstream_inference_cost,
+              createdAt: created_at,
+              isByok: is_byok,
+              providerName: provider_name,
+              finishReason: finish_reason,
+              generationTime: generation_time,
+              promptTokens: native_tokens_prompt,
+              completionTokens: native_tokens_completion,
+              reasoningTokens: native_tokens_reasoning,
+              cachedTokens: native_tokens_cached,
+              cacheCreationTokens: native_tokens_cache_creation,
+              billableWebSearchCalls: billable_web_search_calls,
+            }),
+          ),
+      })
+      .transform(({ data }) => data),
+  ),
+);

package/src/gateway-headers.ts

@@ -0,0 +1,3 @@
+export const GATEWAY_AUTH_METHOD_HEADER = 'ai-gateway-auth-method' as const;
+export const VERCEL_AI_GATEWAY_TEAM_HEADER =
+  'x-vercel-ai-gateway-team' as const;

package/src/gateway-image-model-settings.ts

@@ -1,19 +1,32 @@
 export 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'
-  | 'xai/grok-imagine-image-pro'
   | (string & {});

package/src/gateway-image-model.ts

@@ -1,7 +1,7 @@
 import type {
-  ImageModelV3,
-  ImageModelV3File,
-  ImageModelV3ProviderMetadata,
+  ImageModelV4,
+  ImageModelV4File,
+  ImageModelV4ProviderMetadata,
 } from '@ai-sdk/provider';
 import {
   combineHeaders,
@@ -10,6 +10,9 @@ import {
   createJsonErrorResponseHandler,
   postJsonToApi,
   resolve,
+  serializeModelOptions,
+  WORKFLOW_SERIALIZE,
+  WORKFLOW_DESERIALIZE,
   type Resolvable,
 } from '@ai-sdk/provider-utils';
 import { z } from 'zod/v4';
@@ -17,17 +20,33 @@ import type { GatewayConfig } from './gateway-config';
 import { asGatewayError } from './errors';
 import { parseAuthMethod } from './errors/parse-auth-method';
 
-export class GatewayImageModel implements ImageModelV3 {
-  readonly specificationVersion = 'v3' as const;
+type GatewayImageConfig = GatewayConfig & {
+  provider: string;
+  o11yHeaders: Resolvable<Record<string, string>>;
+};
+
+export class GatewayImageModel implements ImageModelV4 {
+  readonly specificationVersion = 'v4' as const;
   // Set a very large number to prevent client-side splitting of requests
   readonly maxImagesPerCall = Number.MAX_SAFE_INTEGER;
 
+  static [WORKFLOW_SERIALIZE](model: GatewayImageModel) {
+    return serializeModelOptions({
+      modelId: model.modelId,
+      config: model.config,
+    });
+  }
+
+  static [WORKFLOW_DESERIALIZE](options: {
+    modelId: string;
+    config: GatewayImageConfig;
+  }) {
+    return new GatewayImageModel(options.modelId, options.config);
+  }
+
   constructor(
     readonly modelId: string,
-    private readonly config: GatewayConfig & {
-      provider: string;
-      o11yHeaders: Resolvable<Record<string, string>>;
-    },
+    private readonly config: GatewayImageConfig,
   ) {}
 
   get provider(): string {
@@ -45,16 +64,14 @@ export class GatewayImageModel implements ImageModelV3 {
     providerOptions,
     headers,
     abortSignal,
-  }: Parameters<ImageModelV3['doGenerate']>[0]): Promise<
-    Awaited<ReturnType<ImageModelV3['doGenerate']>>
+  }: Parameters<ImageModelV4['doGenerate']>[0]): Promise<
+    Awaited<ReturnType<ImageModelV4['doGenerate']>>
   > {
-    const resolvedHeaders = await resolve(this.config.headers());
+    const resolvedHeaders = this.config.headers
+      ? await resolve(this.config.headers)
+      : undefined;
     try {
-      const {
-        responseHeaders,
-        value: responseBody,
-        rawValue,
-      } = await postJsonToApi({
+      const { responseHeaders, value: responseBody } = await postJsonToApi({
         url: this.getUrl(),
         headers: combineHeaders(
           resolvedHeaders,
@@ -89,7 +106,7 @@ export class GatewayImageModel implements ImageModelV3 {
         images: responseBody.images, // Always base64 strings from server
         warnings: responseBody.warnings ?? [],
         providerMetadata:
-          responseBody.providerMetadata as ImageModelV3ProviderMetadata,
+          responseBody.providerMetadata as ImageModelV4ProviderMetadata,
         response: {
           timestamp: new Date(),
           modelId: this.modelId,
@@ -104,7 +121,10 @@ export class GatewayImageModel implements ImageModelV3 {
         }),
       };
     } catch (error) {
-      throw await asGatewayError(error, await parseAuthMethod(resolvedHeaders));
+      throw await asGatewayError(
+        error,
+        await parseAuthMethod(resolvedHeaders ?? {}),
+      );
     }
   }
 
@@ -114,13 +134,13 @@ export class GatewayImageModel implements ImageModelV3 {
 
   private getModelConfigHeaders() {
     return {
-      'ai-image-model-specification-version': '3',
+      'ai-image-model-specification-version': '4',
       'ai-model-id': this.modelId,
     };
   }
 }
 
-function maybeEncodeImageFile(file: ImageModelV3File) {
+function maybeEncodeImageFile(file: ImageModelV4File) {
   if (file.type === 'file' && file.data instanceof Uint8Array) {
     return {
       ...file,
@@ -147,6 +167,11 @@ const gatewayImageWarningSchema = z.discriminatedUnion('type', [
     feature: z.string(),
     details: z.string().optional(),
   }),
+  z.object({
+    type: z.literal('deprecated'),
+    setting: z.string(),
+    message: z.string(),
+  }),
   z.object({
     type: z.literal('other'),
     message: z.string(),

package/src/gateway-language-model-settings.ts

@@ -3,6 +3,7 @@ export type GatewayModelId =
   | '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'
@@ -13,29 +14,33 @@ export type GatewayModelId =
   | '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-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-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'
@@ -46,25 +51,30 @@ export type GatewayModelId =
   | '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'
+  | '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-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'
   | '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'
+  | 'meituan/longcat-flash-thinking-2601'
   | 'meta/llama-3.1-70b'
   | 'meta/llama-3.1-8b'
   | 'meta/llama-3.2-11b'
@@ -78,6 +88,10 @@ export type GatewayModelId =
   | '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'
@@ -89,23 +103,24 @@ export type GatewayModelId =
   | 'mistral/ministral-8b'
   | 'mistral/mistral-large-3'
   | 'mistral/mistral-medium'
+  | 'mistral/mistral-medium-3.5'
   | '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'
+  | '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/codex-mini'
   | 'openai/gpt-3.5-turbo'
   | 'openai/gpt-3.5-turbo-instruct'
   | 'openai/gpt-4-turbo'
@@ -130,12 +145,14 @@ export type GatewayModelId =
   | '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-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'
@@ -147,23 +164,23 @@ export type GatewayModelId =
   | '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'
+  | '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-code-fast-1'
+  | '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'
@@ -171,6 +188,10 @@ export type GatewayModelId =
   | '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 & {});