feat(provider): add Vercel AI Gateway support

src/cli/cli-installer.telemetry.test.ts

@@ -22,6 +22,7 @@ describe("runCliInstaller telemetry isolation", () => {
         hasZaiCodingPlan: false,
         hasKimiForCoding: false,
         hasOpencodeGo: false,
+        hasVercelAiGateway: false,
       }),
       spyOn(configManager, "isOpenCodeInstalled").mockResolvedValue(true),
       spyOn(configManager, "getOpenCodeVersion").mockResolvedValue("1.4.0"),

src/cli/cli-installer.test.ts

@@ -37,6 +37,7 @@ describe("runCliInstaller", () => {
         hasZaiCodingPlan: false,
         hasKimiForCoding: false,
         hasOpencodeGo: false,
+        hasVercelAiGateway: false,
       }),
       spyOn(configManager, "isOpenCodeInstalled").mockResolvedValue(true),
       spyOn(configManager, "getOpenCodeVersion").mockResolvedValue("1.3.9"),
@@ -83,6 +84,7 @@ describe("runCliInstaller", () => {
         hasZaiCodingPlan: false,
         hasKimiForCoding: false,
         hasOpencodeGo: false,
+        hasVercelAiGateway: false,
       }),
       spyOn(configManager, "isOpenCodeInstalled").mockResolvedValue(true),
       spyOn(configManager, "getOpenCodeVersion").mockResolvedValue("1.4.0"),

src/cli/cli-installer.ts

@@ -138,7 +138,8 @@ export async function runCliInstaller(args: InstallArgs, version: string): Promi
     !config.hasOpenAI &&
     !config.hasGemini &&
     !config.hasCopilot &&
-    !config.hasOpencodeZen
+    !config.hasOpencodeZen &&
+    !config.hasVercelAiGateway
   ) {
     printWarning("No model providers configured. Using opencode/big-pickle as fallback.")
   }

src/cli/cli-program.ts

@@ -33,21 +33,23 @@ program
   .option("--zai-coding-plan <value>", "Z.ai Coding Plan subscription: no, yes (default: no)")
   .option("--kimi-for-coding <value>", "Kimi For Coding subscription: no, yes (default: no)")
   .option("--opencode-go <value>", "OpenCode Go subscription: no, yes (default: no)")
+  .option("--vercel-ai-gateway <value>", "Vercel AI Gateway: no, yes (default: no)")
   .option("--skip-auth", "Skip authentication setup hints")
   .addHelpText("after", `
 Examples:
   $ bunx oh-my-opencode install
   $ bunx oh-my-opencode install --no-tui --claude=max20 --openai=yes --gemini=yes --copilot=no
   $ bunx oh-my-opencode install --no-tui --claude=no --gemini=no --copilot=yes --opencode-zen=yes
 
-Model Providers (Priority: Native > Copilot > OpenCode Zen > Z.ai > Kimi):
+Model Providers (Priority: Native > Copilot > OpenCode Zen > Z.ai > Kimi > Vercel):
   Claude        Native anthropic/ models (Opus, Sonnet, Haiku)
   OpenAI        Native openai/ models (GPT-5.4 for Oracle)
   Gemini        Native google/ models (Gemini 3.1 Pro, Flash)
   Copilot       github-copilot/ models (fallback)
   OpenCode Zen  opencode/ models (opencode/claude-opus-4-6, etc.)
-   Z.ai          zai-coding-plan/glm-5 (visual-engineering fallback)
+  Z.ai          zai-coding-plan/glm-5 (visual-engineering fallback)
   Kimi          kimi-for-coding/k2p5 (Sisyphus/Prometheus fallback)
+  Vercel        vercel/ models (universal proxy, always last fallback)
 `)
   .action(async (options) => {
     const args: InstallArgs = {
@@ -60,6 +62,7 @@ Model Providers (Priority: Native > Copilot > OpenCode Zen > Z.ai > Kimi):
       zaiCodingPlan: options.zaiCodingPlan,
       kimiForCoding: options.kimiForCoding,
       opencodeGo: options.opencodeGo,
+      vercelAiGateway: options.vercelAiGateway,
       skipAuth: options.skipAuth ?? false,
     }
     const exitCode = await install(args)

src/cli/config-manager/detect-current-config.ts

@@ -12,6 +12,7 @@ function detectProvidersFromOmoConfig(): {
   hasZaiCodingPlan: boolean
   hasKimiForCoding: boolean
   hasOpencodeGo: boolean
+  hasVercelAiGateway: boolean
 } {
   const omoConfigPath = getOmoConfigPath()
   if (!existsSync(omoConfigPath)) {
@@ -21,6 +22,7 @@ function detectProvidersFromOmoConfig(): {
       hasZaiCodingPlan: false,
       hasKimiForCoding: false,
       hasOpencodeGo: false,
+      hasVercelAiGateway: false,
     }
   }
 
@@ -34,6 +36,7 @@ function detectProvidersFromOmoConfig(): {
         hasZaiCodingPlan: false,
         hasKimiForCoding: false,
         hasOpencodeGo: false,
+        hasVercelAiGateway: false,
       }
     }
 
@@ -43,15 +46,17 @@ function detectProvidersFromOmoConfig(): {
     const hasZaiCodingPlan = configStr.includes('"zai-coding-plan/')
     const hasKimiForCoding = configStr.includes('"kimi-for-coding/')
     const hasOpencodeGo = configStr.includes('"opencode-go/')
+    const hasVercelAiGateway = configStr.includes('"vercel/')
 
-    return { hasOpenAI, hasOpencodeZen, hasZaiCodingPlan, hasKimiForCoding, hasOpencodeGo }
+    return { hasOpenAI, hasOpencodeZen, hasZaiCodingPlan, hasKimiForCoding, hasOpencodeGo, hasVercelAiGateway }
   } catch {
     return {
       hasOpenAI: true,
       hasOpencodeZen: true,
       hasZaiCodingPlan: false,
       hasKimiForCoding: false,
       hasOpencodeGo: false,
+      hasVercelAiGateway: false,
     }
   }
 }
@@ -78,6 +83,7 @@ export function detectCurrentConfig(): DetectedConfig {
     hasZaiCodingPlan: false,
     hasKimiForCoding: false,
     hasOpencodeGo: false,
+    hasVercelAiGateway: false,
   }
 
   const { format, path } = detectConfigFormat()
@@ -106,12 +112,13 @@ export function detectCurrentConfig(): DetectedConfig {
   const providers = openCodeConfig.provider as Record<string, unknown> | undefined
   result.hasGemini = providers ? "google" in providers : false
 
-  const { hasOpenAI, hasOpencodeZen, hasZaiCodingPlan, hasKimiForCoding, hasOpencodeGo } = detectProvidersFromOmoConfig()
+  const { hasOpenAI, hasOpencodeZen, hasZaiCodingPlan, hasKimiForCoding, hasOpencodeGo, hasVercelAiGateway } = detectProvidersFromOmoConfig()
   result.hasOpenAI = hasOpenAI
   result.hasOpencodeZen = hasOpencodeZen
   result.hasZaiCodingPlan = hasZaiCodingPlan
   result.hasKimiForCoding = hasKimiForCoding
   result.hasOpencodeGo = hasOpencodeGo
+  result.hasVercelAiGateway = hasVercelAiGateway
 
   return result
 }

src/cli/config-manager/generate-omo-config.test.ts

@@ -18,6 +18,7 @@ describe("generateOmoConfig - model fallback system", () => {
       hasZaiCodingPlan: false,
       hasKimiForCoding: false,
       hasOpencodeGo: false,
+      hasVercelAiGateway: false,
     }
 
     //#when
@@ -42,6 +43,7 @@ describe("generateOmoConfig - model fallback system", () => {
       hasZaiCodingPlan: false,
       hasKimiForCoding: false,
       hasOpencodeGo: false,
+      hasVercelAiGateway: false,
     }
 
     //#when
@@ -64,14 +66,15 @@ describe("generateOmoConfig - model fallback system", () => {
       hasZaiCodingPlan: true,
       hasKimiForCoding: false,
       hasOpencodeGo: false,
+      hasVercelAiGateway: false,
     }
 
     //#when
     const result = generateOmoConfig(config)
 
     //#then
     expect((result.agents as Record<string, { model: string }>).librarian.model).toBe("zai-coding-plan/glm-4.7")
-    expect((result.agents as Record<string, { model: string }>).sisyphus.model).toBe("anthropic/claude-opus-4-6")
+    expect((result.agents as Record<string, { model: string }>).sisyphus.model).toBe("anthropic/claude-opus-4.6")
   })
 
   test("uses native OpenAI models when only ChatGPT available", () => {
@@ -86,6 +89,7 @@ describe("generateOmoConfig - model fallback system", () => {
       hasZaiCodingPlan: false,
       hasKimiForCoding: false,
       hasOpencodeGo: false,
+      hasVercelAiGateway: false,
     }
 
     //#when
@@ -110,6 +114,7 @@ describe("generateOmoConfig - model fallback system", () => {
       hasZaiCodingPlan: false,
       hasKimiForCoding: false,
       hasOpencodeGo: false,
+      hasVercelAiGateway: false,
     }
 
     //#when
@@ -126,7 +131,7 @@ describe("generateOmoConfig - model fallback system", () => {
     }>
 
     //#then
-    expect(agents.sisyphus.model).toBe("anthropic/claude-opus-4-6")
+    expect(agents.sisyphus.model).toBe("anthropic/claude-opus-4.6")
     expect(agents.sisyphus.fallback_models).toEqual([
       {
         model: "openai/gpt-5.4",
@@ -136,7 +141,7 @@ describe("generateOmoConfig - model fallback system", () => {
     expect(categories.deep.model).toBe("openai/gpt-5.4")
     expect(categories.deep.fallback_models).toEqual([
       {
-        model: "anthropic/claude-opus-4-6",
+        model: "anthropic/claude-opus-4.6",
         variant: "max",
       },
     ])
@@ -154,6 +159,7 @@ describe("generateOmoConfig - model fallback system", () => {
       hasZaiCodingPlan: false,
       hasKimiForCoding: false,
       hasOpencodeGo: false,
+      hasVercelAiGateway: false,
     }
 
     //#when
@@ -175,6 +181,7 @@ describe("generateOmoConfig - model fallback system", () => {
       hasZaiCodingPlan: false,
       hasKimiForCoding: false,
       hasOpencodeGo: false,
+      hasVercelAiGateway: false,
     }
 
     //#when

src/cli/config-manager/write-omo-config.test.ts

@@ -20,6 +20,7 @@ const installConfig: InstallConfig = {
   hasZaiCodingPlan: false,
   hasKimiForCoding: false,
   hasOpencodeGo: false,
+  hasVercelAiGateway: false,
 }
 
 function getRecord(value: unknown): Record<string, unknown> {

src/cli/install-validators.ts

@@ -40,6 +40,7 @@ export function formatConfigSummary(config: InstallConfig): string {
   lines.push(formatProvider("OpenCode Zen", config.hasOpencodeZen, "opencode/ models"))
   lines.push(formatProvider("Z.ai Coding Plan", config.hasZaiCodingPlan, "Librarian/Multimodal"))
   lines.push(formatProvider("Kimi For Coding", config.hasKimiForCoding, "Sisyphus/Prometheus fallback"))
+  lines.push(formatProvider("Vercel AI Gateway", config.hasVercelAiGateway, "universal proxy"))
 
   lines.push("")
   lines.push(color.dim("─".repeat(40)))
@@ -153,6 +154,10 @@ export function validateNonTuiArgs(args: InstallArgs): { valid: boolean; errors:
     errors.push(`Invalid --kimi-for-coding value: ${args.kimiForCoding} (expected: no, yes)`)
   }
 
+  if (args.vercelAiGateway !== undefined && !["no", "yes"].includes(args.vercelAiGateway)) {
+    errors.push(`Invalid --vercel-ai-gateway value: ${args.vercelAiGateway} (expected: no, yes)`)
+  }
+
   return { valid: errors.length === 0, errors }
 }
 
@@ -167,6 +172,7 @@ export function argsToConfig(args: InstallArgs): InstallConfig {
     hasZaiCodingPlan: args.zaiCodingPlan === "yes",
 hasKimiForCoding: args.kimiForCoding === "yes",
     hasOpencodeGo: args.opencodeGo === "yes",
+    hasVercelAiGateway: args.vercelAiGateway === "yes",
   }
 }
 
@@ -179,6 +185,7 @@ export function detectedToInitialValues(detected: DetectedConfig): {
   zaiCodingPlan: BooleanArg
 kimiForCoding: BooleanArg
   opencodeGo: BooleanArg
+  vercelAiGateway: BooleanArg
 } {
   let claude: ClaudeSubscription = "no"
   if (detected.hasClaude) {
@@ -194,5 +201,6 @@ kimiForCoding: BooleanArg
     zaiCodingPlan: detected.hasZaiCodingPlan ? "yes" : "no",
 kimiForCoding: detected.hasKimiForCoding ? "yes" : "no",
     opencodeGo: detected.hasOpencodeGo ? "yes" : "no",
+    vercelAiGateway: detected.hasVercelAiGateway ? "yes" : "no",
   }
 }

src/cli/model-fallback-types.ts

@@ -11,6 +11,7 @@ export interface ProviderAvailability {
 	zai: boolean
 kimiForCoding: boolean
 	opencodeGo: boolean
+	vercelAiGateway: boolean
 	isMaxPlan: boolean
 }
 

src/cli/model-fallback.test.ts

@@ -16,6 +16,7 @@ function createConfig(overrides: Partial<InstallConfig> = {}): InstallConfig {
     hasZaiCodingPlan: false,
     hasKimiForCoding: false,
     hasOpencodeGo: false,
+    hasVercelAiGateway: false,
     ...overrides,
   }
 }
@@ -607,6 +608,74 @@ describe("generateModelConfig", () => {
     })
   })
 
+  describe("Vercel AI Gateway provider", () => {
+    test("uses vercel/ model strings when only Vercel AI Gateway is available", () => {
+      // #given only Vercel AI Gateway is available
+      const config = createConfig({ hasVercelAiGateway: true })
+
+      // #when generateModelConfig is called
+      const result = generateModelConfig(config)
+
+      // #then should use vercel/<sub-provider>/<model> format
+      expect(result).toMatchSnapshot()
+    })
+
+    test("uses vercel/ model strings with isMax20 flag", () => {
+      // #given Vercel AI Gateway is available with Max 20 plan
+      const config = createConfig({ hasVercelAiGateway: true, isMax20: true })
+
+      // #when generateModelConfig is called
+      const result = generateModelConfig(config)
+
+      // #then should use higher capability models via gateway
+      expect(result).toMatchSnapshot()
+    })
+
+    test("explore uses vercel/minimax/minimax-m2.7-highspeed when only gateway available", () => {
+      // #given only Vercel AI Gateway is available
+      const config = createConfig({ hasVercelAiGateway: true })
+
+      // #when generateModelConfig is called
+      const result = generateModelConfig(config)
+
+      // #then explore should use gateway-routed minimax (preferred over claude-haiku)
+      expect(result.agents?.explore?.model).toBe("vercel/minimax/minimax-m2.7-highspeed")
+    })
+
+    test("librarian uses vercel/minimax/minimax-m2.7 when only gateway available", () => {
+      // #given only Vercel AI Gateway is available
+      const config = createConfig({ hasVercelAiGateway: true })
+
+      // #when generateModelConfig is called
+      const result = generateModelConfig(config)
+
+      // #then librarian should use gateway-routed minimax (preferred over claude-haiku)
+      expect(result.agents?.librarian?.model).toBe("vercel/minimax/minimax-m2.7")
+    })
+
+    test("Hephaestus is created when only Vercel AI Gateway is available", () => {
+      // #given only Vercel AI Gateway is available
+      const config = createConfig({ hasVercelAiGateway: true })
+
+      // #when generateModelConfig is called
+      const result = generateModelConfig(config)
+
+      // #then hephaestus should be created with gateway-routed gpt-5.4
+      expect(result.agents?.hephaestus?.model).toBe("vercel/openai/gpt-5.4")
+    })
+
+    test("native providers take priority over gateway", () => {
+      // #given Claude and Vercel AI Gateway are both available
+      const config = createConfig({ hasClaude: true, hasVercelAiGateway: true })
+
+      // #when generateModelConfig is called
+      const result = generateModelConfig(config)
+
+      // #then should prefer native anthropic over gateway
+      expect(result.agents?.sisyphus?.model).toBe("anthropic/claude-opus-4.6")
+    })
+  })
+
   describe("schema URL", () => {
     test("always includes correct schema URL", () => {
       // #given any config

src/cli/model-fallback.ts

@@ -105,7 +105,8 @@ export function generateModelConfig(config: InstallConfig): GeneratedOmoConfig {
     avail.copilot ||
     avail.zai ||
     avail.kimiForCoding ||
-    avail.opencodeGo
+    avail.opencodeGo ||
+    avail.vercelAiGateway
   if (!hasAnyProvider) {
     return {
       $schema: SCHEMA_URL,
@@ -130,6 +131,8 @@ export function generateModelConfig(config: InstallConfig): GeneratedOmoConfig {
         agentConfig = { model: "opencode-go/minimax-m2.7" }
       } else if (avail.zai) {
         agentConfig = { model: ZAI_MODEL }
+      } else if (avail.vercelAiGateway) {
+        agentConfig = { model: "vercel/minimax/minimax-m2.7" }
       }
       if (agentConfig) {
         agents[role] = attachAllFallbackModels(agentConfig, req.fallbackChain, avail)
@@ -147,6 +150,8 @@ export function generateModelConfig(config: InstallConfig): GeneratedOmoConfig {
         agentConfig = { model: "opencode-go/minimax-m2.7" }
       } else if (avail.copilot) {
         agentConfig = { model: "github-copilot/gpt-5-mini" }
+      } else if (avail.vercelAiGateway) {
+        agentConfig = { model: "vercel/minimax/minimax-m2.7-highspeed" }
       } else {
         agentConfig = { model: "opencode/gpt-5-nano" }
       }

src/cli/openai-only-model-catalog.test.ts

@@ -14,6 +14,7 @@ function createConfig(overrides: Partial<InstallConfig> = {}): InstallConfig {
     hasZaiCodingPlan: false,
     hasKimiForCoding: false,
     hasOpencodeGo: false,
+    hasVercelAiGateway: false,
     ...overrides,
   }
 }