diff --git a/packages/app/server/src/env.ts b/packages/app/server/src/env.ts
index da23dd994..5933baeb7 100644
--- a/packages/app/server/src/env.ts
+++ b/packages/app/server/src/env.ts
@@ -45,6 +45,8 @@ export const env = createEnv({
     GROQ_API_KEY: z.string().optional(),
     XAI_API_KEY: z.string().optional(),
     OPENROUTER_API_KEY: z.string().optional(),
+    VERCEL_GATEWAY_API_KEY: z.string().optional(),
+    VERCEL_GATEWAY_BASE_URL: z.string().url().optional(),
     TAVILY_API_KEY: z.string().optional(),
     E2B_API_KEY: z.string().optional(),
     GOOGLE_SERVICE_ACCOUNT_KEY_ENCODED: z.string().optional(),
diff --git a/packages/app/server/src/providers/ProviderFactory.ts b/packages/app/server/src/providers/ProviderFactory.ts
index 8bc01e719..7c40883d5 100644
--- a/packages/app/server/src/providers/ProviderFactory.ts
+++ b/packages/app/server/src/providers/ProviderFactory.ts
@@ -22,6 +22,7 @@ import { OpenAIResponsesProvider } from './OpenAIResponsesProvider';
 import { OpenRouterProvider } from './OpenRouterProvider';
 import { ProviderType } from './ProviderType';
 import { XAIProvider } from './XAIProvider';
+import { VercelGatewayProvider } from './VercelGatewayProvider';
 import {
   VertexAIProvider,
   PROXY_PASSTHROUGH_ONLY_MODEL as VertexAIProxyPassthroughOnlyModel,
@@ -58,6 +59,9 @@ const createChatModelToProviderMapping = (): Record<string, ProviderType> => {
         case 'Xai':
           mapping[modelConfig.model_id] = ProviderType.XAI;
           break;
+        case 'VercelGateway':
+          mapping[modelConfig.model_id] = ProviderType.VERCEL_GATEWAY;
+          break;
         // Add other providers as needed
         default:
           // Skip models with unsupported providers
@@ -192,6 +196,8 @@ export const getProvider = (
       return new GroqProvider(stream, model);
     case ProviderType.XAI:
       return new XAIProvider(stream, model);
+    case ProviderType.VERCEL_GATEWAY:
+      return new VercelGatewayProvider(stream, model);
     default:
       throw new Error(`Unknown provider type: ${type}`);
   }
diff --git a/packages/app/server/src/providers/ProviderType.ts b/packages/app/server/src/providers/ProviderType.ts
index b2514ac80..93ab8c6aa 100644
--- a/packages/app/server/src/providers/ProviderType.ts
+++ b/packages/app/server/src/providers/ProviderType.ts
@@ -12,4 +12,5 @@ export enum ProviderType {
   OPENAI_VIDEOS = 'OPENAI_VIDEOS',
   GROQ = 'GROQ',
   XAI = 'XAI',
+  VERCEL_GATEWAY = 'VERCEL_GATEWAY',
 }
diff --git a/packages/app/server/src/providers/VercelGatewayProvider.ts b/packages/app/server/src/providers/VercelGatewayProvider.ts
new file mode 100644
index 000000000..459b4d254
--- /dev/null
+++ b/packages/app/server/src/providers/VercelGatewayProvider.ts
@@ -0,0 +1,117 @@
+import { LlmTransactionMetadata, Transaction } from '../types';
+import { getCostPerToken } from '../services/AccountingService';
+import { BaseProvider } from './BaseProvider';
+import { ProviderType } from './ProviderType';
+import logger from '../logger';
+import { env } from '../env';
+
+interface CompletionStateBody {
+  id: string;
+  usage: {
+    prompt_tokens: number;
+    completion_tokens: number;
+    total_tokens: number;
+  };
+}
+
+interface StreamingChunkBody {
+  id: string;
+  choices: {
+    index: number;
+    delta: {
+      content?: string;
+    };
+    finish_reason: string | null;
+  }[];
+  usage: {
+    prompt_tokens: number;
+    completion_tokens: number;
+    total_tokens: number;
+  } | null;
+}
+
+const parseSSEGPTFormat = (data: string): StreamingChunkBody[] => {
+  const events = data.split('\n\n');
+  const chunks: StreamingChunkBody[] = [];
+
+  for (const event of events) {
+    if (!event.trim()) continue;
+    if (event.startsWith('data: ')) {
+      const jsonStr = event.slice(6);
+      if (jsonStr.trim() === '[DONE]') continue;
+      try {
+        const parsed = JSON.parse(jsonStr);
+        chunks.push(parsed);
+      } catch (error) {
+        logger.error(`Error parsing SSE chunk: ${error}`);
+      }
+    }
+  }
+
+  return chunks;
+};
+
+export class VercelGatewayProvider extends BaseProvider {
+  getType(): ProviderType {
+    return ProviderType.VERCEL_GATEWAY;
+  }
+
+  getBaseUrl(): string {
+    return env.VERCEL_GATEWAY_BASE_URL || 'https://ai-gateway.vercel.sh/v1/ai';
+  }
+
+  getApiKey(): string | undefined {
+    return env.VERCEL_GATEWAY_API_KEY;
+  }
+
+  async handleBody(data: string): Promise<Transaction> {
+    try {
+      let prompt_tokens = 0;
+      let completion_tokens = 0;
+      let total_tokens = 0;
+      let providerId = 'null';
+
+      if (this.getIsStream()) {
+        const chunks = parseSSEGPTFormat(data);
+        for (const chunk of chunks) {
+          if (chunk.usage !== null) {
+            prompt_tokens += chunk.usage.prompt_tokens;
+            completion_tokens += chunk.usage.completion_tokens;
+            total_tokens += chunk.usage.total_tokens;
+          }
+          providerId = chunk.id || 'null';
+        }
+      } else {
+        const parsed = JSON.parse(data) as CompletionStateBody;
+        prompt_tokens += parsed.usage.prompt_tokens;
+        completion_tokens += parsed.usage.completion_tokens;
+        total_tokens += parsed.usage.total_tokens;
+        providerId = parsed.id || 'null';
+      }
+
+      const cost = getCostPerToken(
+        this.getModel(),
+        prompt_tokens,
+        completion_tokens
+      );
+
+      const metadata: LlmTransactionMetadata = {
+        providerId: providerId,
+        provider: this.getType(),
+        model: this.getModel(),
+        inputTokens: prompt_tokens,
+        outputTokens: completion_tokens,
+        totalTokens: total_tokens,
+      };
+
+      return {
+        rawTransactionCost: cost,
+        metadata: metadata,
+        status: 'success',
+      };
+    } catch (error) {
+      logger.error(`Error processing data: ${error}`);
+      throw error;
+    }
+  }
+}
diff --git a/packages/app/server/src/services/AccountingService.ts b/packages/app/server/src/services/AccountingService.ts
index 02e51e14e..34c3abcab 100644
--- a/packages/app/server/src/services/AccountingService.ts
+++ b/packages/app/server/src/services/AccountingService.ts
@@ -10,6 +10,7 @@ import {
   SupportedImageModel,
   SupportedVideoModel,
   XAIModels,
+  VercelGatewayModels,
 } from '@merit-systems/echo-typescript-sdk';
 
 import { Decimal } from '@prisma/client/runtime/library';
@@ -30,6 +31,7 @@ export const ALL_SUPPORTED_MODELS: SupportedModel[] = [
   ...OpenRouterModels,
   ...GroqModels,
   ...XAIModels,
+  ...VercelGatewayModels,
 ];
 
 // Handle image models separately since they have different pricing structure
diff --git a/packages/sdk/next/src/ai-providers/index.ts b/packages/sdk/next/src/ai-providers/index.ts
index 4ea2555df..c938a42da 100644
--- a/packages/sdk/next/src/ai-providers/index.ts
+++ b/packages/sdk/next/src/ai-providers/index.ts
@@ -3,3 +3,4 @@ export * from './google';
 export * from './xai';
 export * from './groq';
 export * from './openai';
+export * from './vercel';
diff --git a/packages/sdk/next/src/ai-providers/vercel.ts b/packages/sdk/next/src/ai-providers/vercel.ts
new file mode 100644
index 000000000..60974288b
--- /dev/null
+++ b/packages/sdk/next/src/ai-providers/vercel.ts
@@ -0,0 +1,10 @@
+import { getEchoToken } from '../auth/token-manager';
+import {
+  createEchoVercelGateway as createEchoVercelGatewayBase,
+  EchoConfig,
+  GatewayProvider,
+} from '@merit-systems/echo-typescript-sdk';
+
+export function createEchoVercelGateway(config: EchoConfig): GatewayProvider {
+  return createEchoVercelGatewayBase(config, async () => getEchoToken(config));
+}
diff --git a/packages/sdk/ts/package.json b/packages/sdk/ts/package.json
index f2a7f1066..43e5388d2 100644
--- a/packages/sdk/ts/package.json
+++ b/packages/sdk/ts/package.json
@@ -63,6 +63,7 @@
     "@ai-sdk/openai": "2.0.32",
     "@ai-sdk/xai": "2.0.16",
     "@openrouter/ai-sdk-provider": "1.2.0",
+    "@ai-sdk/gateway": "^1.0.12",
     "ai": "5.0.47"
   }
 }
diff --git a/packages/sdk/ts/src/__tests__/vercel-gateway-models.test.ts b/packages/sdk/ts/src/__tests__/vercel-gateway-models.test.ts
new file mode 100644
index 000000000..3c107d42f
--- /dev/null
+++ b/packages/sdk/ts/src/__tests__/vercel-gateway-models.test.ts
@@ -0,0 +1,45 @@
+import { describe, it, expect } from 'vitest';
+import { VercelGatewayModels } from '../src/supported-models/chat/vercel-gateway';
+
+describe('VercelGatewayModels', () => {
+  it('should export a non-empty array of models', () => {
+    expect(VercelGatewayModels.length).toBeGreaterThan(0);
+  });
+
+  it('should have all models with VercelGateway provider', () => {
+    for (const model of VercelGatewayModels) {
+      expect(model.provider).toBe('VercelGateway');
+    }
+  });
+
+  it('should have valid pricing for all models', () => {
+    for (const model of VercelGatewayModels) {
+      expect(model.input_cost_per_token).toBeGreaterThan(0);
+      expect(model.output_cost_per_token).toBeGreaterThan(0);
+    }
+  });
+
+  it('should have prefixed model IDs (provider/model-name)', () => {
+    for (const model of VercelGatewayModels) {
+      expect(model.model_id).toMatch(/^[a-z]+\/[a-z0-9._-]+$/);
+    }
+  });
+
+  it('should include key OpenAI models', () => {
+    const ids = VercelGatewayModels.map(m => m.model_id);
+    expect(ids).toContain('openai/gpt-4o');
+    expect(ids).toContain('openai/gpt-4o-mini');
+  });
+
+  it('should include key Anthropic models', () => {
+    const ids = VercelGatewayModels.map(m => m.model_id);
+    expect(ids).toContain('anthropic/claude-sonnet-4');
+    expect(ids).toContain('anthropic/claude-3.5-sonnet');
+  });
+
+  it('should include key Google models', () => {
+    const ids = VercelGatewayModels.map(m => m.model_id);
+    expect(ids).toContain('google/gemini-2.5-flash');
+    expect(ids).toContain('google/gemini-2.5-pro');
+  });
+});
diff --git a/packages/sdk/ts/src/__tests__/vercel-gateway.test.ts b/packages/sdk/ts/src/__tests__/vercel-gateway.test.ts
new file mode 100644
index 000000000..02e3f0e92
--- /dev/null
+++ b/packages/sdk/ts/src/__tests__/vercel-gateway.test.ts
@@ -0,0 +1,66 @@
+import { describe, it, expect, vi } from 'vitest';
+import { createEchoVercelGateway } from '../src/providers/vercel';
+import { echoFetch } from '../src/providers/index';
+
+// Mock @ai-sdk/gateway
+vi.mock('@ai-sdk/gateway', () => ({
+  createGatewayProvider: vi.fn((options) => ({
+    _type: 'gateway',
+    _options: options,
+    languageModel: vi.fn(),
+    getAvailableModels: vi.fn(),
+    getCredits: vi.fn(),
+    textEmbeddingModel: vi.fn(),
+  })),
+}));
+
+describe('createEchoVercelGateway', () => {
+  const mockGetTokenFn = vi.fn();
+  const mockOnInsufficientFunds = vi.fn();
+
+  it('should create a gateway provider with echo fetch wrapper', () => {
+    const provider = createEchoVercelGateway(
+      { appId: '60601628-cdb7-481e-8f7e-921981220348' },
+      mockGetTokenFn,
+      mockOnInsufficientFunds
+    );
+
+    expect(provider._type).toBe('gateway');
+    expect(provider._options.apiKey).toBe('placeholder_replaced_by_echoFetch');
+    expect(provider._options.baseURL).toBe('https://echo.router.merit.systems');
+    expect(provider._options.fetch).toBeDefined();
+  });
+
+  it('should use custom baseRouterUrl when provided', () => {
+    const provider = createEchoVercelGateway(
+      {
+        appId: '60601628-cdb7-481e-8f7e-921981220348',
+        baseRouterUrl: 'https://custom-gateway.example.com',
+      },
+      mockGetTokenFn,
+      mockOnInsufficientFunds
+    );
+
+    expect(provider._options.baseURL).toBe('https://custom-gateway.example.com');
+  });
+
+  it('should throw on invalid appId', () => {
+    expect(() =>
+      createEchoVercelGateway(
+        { appId: 'invalid' },
+        mockGetTokenFn,
+        mockOnInsufficientFunds
+      )
+    ).toThrow('Invalid Echo App ID');
+  });
+
+  it('should throw on empty appId', () => {
+    expect(() =>
+      createEchoVercelGateway(
+        { appId: '' },
+        mockGetTokenFn,
+        mockOnInsufficientFunds
+      )
+    ).toThrow('Invalid Echo App ID');
+  });
+});
diff --git a/packages/sdk/ts/src/index.ts b/packages/sdk/ts/src/index.ts
index 521a30f93..0589dd8e7 100644
--- a/packages/sdk/ts/src/index.ts
+++ b/packages/sdk/ts/src/index.ts
@@ -11,6 +11,7 @@ export * from './utils/error-handling';
 export * from './utils/validation';
 export * from './providers';
 export { createEchoXAI } from './providers/xai';
+export { createEchoVercelGateway } from './providers/vercel';
 
 // Export tool-related types and utilities
 export type {
@@ -48,6 +49,7 @@ export type { OpenRouterModel } from './supported-models/chat/openrouter';
 export { GroqModels } from './supported-models/chat/groq';
 export type { GroqModel } from './supported-models/chat/groq';
 export { XAIModels } from './supported-models/chat/xai';
+export { VercelGatewayModels } from './supported-models/chat/vercel-gateway';
 export type { XAIModel } from './supported-models/chat/xai';
 export { OpenAIImageModels } from './supported-models/image/openai';
 export type { OpenAIImageModel } from './supported-models/image/openai';
diff --git a/packages/sdk/ts/src/providers/index.ts b/packages/sdk/ts/src/providers/index.ts
index 62f54fac8..ed07b6453 100644
--- a/packages/sdk/ts/src/providers/index.ts
+++ b/packages/sdk/ts/src/providers/index.ts
@@ -4,6 +4,7 @@ export * from './groq';
 export * from './xai';
 export * from './openai';
 export * from './openrouter';
+export * from './vercel';
 
 export function echoFetch(
   originalFetch: typeof fetch,
@@ -63,3 +64,4 @@ export { type GroqProvider } from '@ai-sdk/groq';
 export { type OpenAIProvider } from '@ai-sdk/openai';
 export { type OpenRouterProvider } from '@openrouter/ai-sdk-provider';
 export { type XaiProvider } from '@ai-sdk/xai';
+export { type GatewayProvider } from '@ai-sdk/gateway';
diff --git a/packages/sdk/ts/src/providers/vercel.ts b/packages/sdk/ts/src/providers/vercel.ts
new file mode 100644
index 000000000..a3250baf4
--- /dev/null
+++ b/packages/sdk/ts/src/providers/vercel.ts
@@ -0,0 +1,26 @@
+import {
+  createGatewayProvider,
+  GatewayProvider,
+} from '@ai-sdk/gateway';
+import { ROUTER_BASE_URL } from 'config';
+import { EchoConfig } from '../types';
+import { validateAppId } from '../utils/validation';
+import { echoFetch } from './index';
+
+export function createEchoVercelGateway(
+  { appId, baseRouterUrl = ROUTER_BASE_URL }: EchoConfig,
+  getTokenFn: (appId: string) => Promise<string | null>,
+  onInsufficientFunds?: () => void
+): GatewayProvider {
+  validateAppId(appId, 'createEchoVercelGateway');
+
+  return createGatewayProvider({
+    baseURL: baseRouterUrl,
+    apiKey: 'placeholder_replaced_by_echoFetch',
+    fetch: echoFetch(
+      fetch,
+      async () => await getTokenFn(appId),
+      onInsufficientFunds
+    ),
+  });
+}
diff --git a/packages/sdk/ts/src/supported-models/chat/vercel-gateway.ts b/packages/sdk/ts/src/supported-models/chat/vercel-gateway.ts
new file mode 100644
index 000000000..7376cf530
--- /dev/null
+++ b/packages/sdk/ts/src/supported-models/chat/vercel-gateway.ts
@@ -0,0 +1,272 @@
+import { SupportedModel } from '../types';
+
+// Union type of all valid Vercel AI Gateway model IDs
+// Models are prefixed with their upstream provider (e.g., openai/gpt-4o, anthropic/claude-3.5-sonnet)
+export type VercelGatewayModel =
+  | 'openai/gpt-4o'
+  | 'openai/gpt-4o-mini'
+  | 'openai/gpt-4.1'
+  | 'openai/gpt-4.1-mini'
+  | 'openai/gpt-4.1-nano'
+  | 'openai/gpt-5'
+  | 'openai/gpt-5-mini'
+  | 'openai/gpt-5-nano'
+  | 'openai/o3'
+  | 'openai/o3-mini'
+  | 'openai/o4-mini'
+  | 'anthropic/claude-3.5-sonnet'
+  | 'anthropic/claude-3.7-sonnet'
+  | 'anthropic/claude-sonnet-4'
+  | 'anthropic/claude-opus-4'
+  | 'anthropic/claude-opus-4.1'
+  | 'anthropic/claude-3-haiku'
+  | 'anthropic/claude-3.5-haiku'
+  | 'anthropic/claude-3-opus'
+  | 'google/gemini-2.5-flash'
+  | 'google/gemini-2.5-flash-lite'
+  | 'google/gemini-2.5-pro'
+  | 'google/gemini-2.0-flash'
+  | 'google/gemini-2.0-flash-lite'
+  | 'deepseek/deepseek-r1'
+  | 'deepseek/deepseek-v3'
+  | 'deepseek/deepseek-v3.1'
+  | 'meta/llama-4-maverick'
+  | 'meta/llama-4-scout'
+  | 'xai/grok-3'
+  | 'xai/grok-3-mini'
+  | 'xai/grok-3-fast'
+  | 'xai/grok-4'
+  | 'mistral/mistral-large'
+  | 'mistral/mistral-medium'
+  | 'mistral/mistral-small'
+  | (string & {});
+
+// Vercel AI Gateway model pricing
+// Source: https://vercel.com/docs/ai-gateway/pricing
+// Note: Vercel Gateway adds its own margin on top of upstream provider costs.
+// These prices reflect Vercel's listed gateway prices.
+export const VercelGatewayModels: SupportedModel[] = [
+  // OpenAI models via Vercel AI Gateway
+  {
+    model_id: 'openai/gpt-4o',
+    input_cost_per_token: 2.5e-6,
+    output_cost_per_token: 1e-5,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'openai/gpt-4o-mini',
+    input_cost_per_token: 1.5e-7,
+    output_cost_per_token: 6e-7,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'openai/gpt-4.1',
+    input_cost_per_token: 2e-6,
+    output_cost_per_token: 8e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'openai/gpt-4.1-mini',
+    input_cost_per_token: 4e-7,
+    output_cost_per_token: 1.6e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'openai/gpt-4.1-nano',
+    input_cost_per_token: 1e-7,
+    output_cost_per_token: 4e-7,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'openai/gpt-5',
+    input_cost_per_token: 1.5e-5,
+    output_cost_per_token: 7.5e-5,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'openai/gpt-5-mini',
+    input_cost_per_token: 1.1e-6,
+    output_cost_per_token: 4.4e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'openai/gpt-5-nano',
+    input_cost_per_token: 5e-7,
+    output_cost_per_token: 2e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'openai/o3',
+    input_cost_per_token: 2e-5,
+    output_cost_per_token: 8e-5,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'openai/o3-mini',
+    input_cost_per_token: 1.1e-6,
+    output_cost_per_token: 4.4e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'openai/o4-mini',
+    input_cost_per_token: 1.1e-6,
+    output_cost_per_token: 4.4e-6,
+    provider: 'VercelGateway',
+  },
+  // Anthropic models via Vercel AI Gateway
+  {
+    model_id: 'anthropic/claude-sonnet-4',
+    input_cost_per_token: 3e-6,
+    output_cost_per_token: 1.5e-5,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'anthropic/claude-opus-4',
+    input_cost_per_token: 1.5e-5,
+    output_cost_per_token: 7.5e-5,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'anthropic/claude-opus-4.1',
+    input_cost_per_token: 1.5e-5,
+    output_cost_per_token: 7.5e-5,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'anthropic/claude-3.7-sonnet',
+    input_cost_per_token: 3e-6,
+    output_cost_per_token: 1.5e-5,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'anthropic/claude-3.5-sonnet',
+    input_cost_per_token: 3e-6,
+    output_cost_per_token: 1.5e-5,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'anthropic/claude-3.5-haiku',
+    input_cost_per_token: 8e-7,
+    output_cost_per_token: 4e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'anthropic/claude-3-haiku',
+    input_cost_per_token: 2.5e-7,
+    output_cost_per_token: 1.25e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'anthropic/claude-3-opus',
+    input_cost_per_token: 1.5e-5,
+    output_cost_per_token: 7.5e-5,
+    provider: 'VercelGateway',
+  },
+  // Google models via Vercel AI Gateway
+  {
+    model_id: 'google/gemini-2.5-pro',
+    input_cost_per_token: 1.25e-6,
+    output_cost_per_token: 1e-5,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'google/gemini-2.5-flash',
+    input_cost_per_token: 1.5e-7,
+    output_cost_per_token: 1.5e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'google/gemini-2.5-flash-lite',
+    input_cost_per_token: 5e-8,
+    output_cost_per_token: 5e-7,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'google/gemini-2.0-flash',
+    input_cost_per_token: 1e-7,
+    output_cost_per_token: 4e-7,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'google/gemini-2.0-flash-lite',
+    input_cost_per_token: 7.5e-8,
+    output_cost_per_token: 3e-7,
+    provider: 'VercelGateway',
+  },
+  // DeepSeek models via Vercel AI Gateway
+  {
+    model_id: 'deepseek/deepseek-r1',
+    input_cost_per_token: 5.5e-6,
+    output_cost_per_token: 2.19e-5,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'deepseek/deepseek-v3',
+    input_cost_per_token: 2.7e-7,
+    output_cost_per_token: 1.1e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'deepseek/deepseek-v3.1',
+    input_cost_per_token: 2.7e-7,
+    output_cost_per_token: 1.1e-6,
+    provider: 'VercelGateway',
+  },
+  // xAI models via Vercel AI Gateway
+  {
+    model_id: 'xai/grok-3',
+    input_cost_per_token: 3e-6,
+    output_cost_per_token: 1.5e-5,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'xai/grok-3-mini',
+    input_cost_per_token: 3e-7,
+    output_cost_per_token: 1.5e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'xai/grok-3-fast',
+    input_cost_per_token: 5e-7,
+    output_cost_per_token: 2.5e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'xai/grok-4',
+    input_cost_per_token: 3e-6,
+    output_cost_per_token: 1.5e-5,
+    provider: 'VercelGateway',
+  },
+  // Meta Llama models via Vercel AI Gateway
+  {
+    model_id: 'meta/llama-4-maverick',
+    input_cost_per_token: 2e-7,
+    output_cost_per_token: 6e-7,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'meta/llama-4-scout',
+    input_cost_per_token: 1e-7,
+    output_cost_per_token: 3e-7,
+    provider: 'VercelGateway',
+  },
+  // Mistral models via Vercel AI Gateway
+  {
+    model_id: 'mistral/mistral-large',
+    input_cost_per_token: 2e-6,
+    output_cost_per_token: 6e-6,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'mistral/mistral-medium',
+    input_cost_per_token: 2.75e-7,
+    output_cost_per_token: 8.25e-7,
+    provider: 'VercelGateway',
+  },
+  {
+    model_id: 'mistral/mistral-small',
+    input_cost_per_token: 1e-7,
+    output_cost_per_token: 3e-7,
+    provider: 'VercelGateway',
+  },
+];
diff --git a/packages/sdk/ts/src/supported-models/index.ts b/packages/sdk/ts/src/supported-models/index.ts
index 3f641501d..b85a35cc9 100644
--- a/packages/sdk/ts/src/supported-models/index.ts
+++ b/packages/sdk/ts/src/supported-models/index.ts
@@ -4,6 +4,7 @@ export * from './chat/anthropic';
 export * from './chat/gemini';
 export * from './chat/openai';
 export * from './chat/openrouter';
+export * from './chat/vercel-gateway';
 export * from './image/openai';
 export * from './responses/openai';
 export * from './video/gemini';