diff --git a/src/shared/models/index.ts b/src/shared/models/index.ts
index 304c6909a5..564cffa006 100644
--- a/src/shared/models/index.ts
+++ b/src/shared/models/index.ts
@@ -27,7 +27,8 @@ export const aiProviderNameHash: Record<ModelProviderEnum, string> = {
   [ModelProviderEnum.LMStudio]: 'LM Studio API',
   [ModelProviderEnum.Perplexity]: 'Perplexity API',
   [ModelProviderEnum.XAI]: 'xAI API',
-  [ModelProviderEnum.OpenRouter]: 'OpenRouter API',
+  [ModelProviderEnum.OpenRouter]: "OpenRouter API",
+  [ModelProviderEnum.Kevoryn]: "Kevoryn API",
   [ModelProviderEnum.Bedrock]: 'AWS Bedrock',
   [ModelProviderEnum.VercelAIGateway]: 'Vercel AI Gateway',
   [ModelProviderEnum.Custom]: 'Custom Provider',
diff --git a/src/shared/providers/definitions/kevoryn.ts b/src/shared/providers/definitions/kevoryn.ts
new file mode 100644
index 0000000000..fd6926d7c2
--- /dev/null
+++ b/src/shared/providers/definitions/kevoryn.ts
@@ -0,0 +1,141 @@
+import { ModelProviderEnum, ModelProviderType } from '../../types'
+import { defineProvider } from '../registry'
+import Kevoryn from './models/kevoryn'
+
+export const kevorynProvider = defineProvider({
+  id: ModelProviderEnum.Kevoryn,
+  name: 'Kevoryn',
+  type: ModelProviderType.OpenAI,
+  description: 'AI API Gateway — 66+ frontier models (GPT, Claude, Gemini, DeepSeek, Kimi, Qwen) through a single endpoint.',
+  urls: {
+    website: 'https://kevoryn.com/',
+    docs: 'https://kevoryn.com/docs',
+    models: 'https://kevoryn.com/models',
+  },
+  defaultSettings: {
+    apiHost: 'https://api.kevoryn.com/v1',
+    models: [
+      {
+        modelId: 'claude-opus-4-8',
+        nickname: 'Claude Opus 4.8',
+        capabilities: ['tool_use', 'reasoning', 'vision'],
+        contextWindow: 200_000,
+        maxOutput: 128_000,
+      },
+      {
+        modelId: 'claude-sonnet-4-6',
+        nickname: 'Claude Sonnet 4.6',
+        capabilities: ['tool_use', 'reasoning', 'vision'],
+        contextWindow: 200_000,
+        maxOutput: 64_000,
+      },
+      {
+        modelId: 'claude-haiku-4-5',
+        nickname: 'Claude Haiku 4.5',
+        capabilities: ['tool_use', 'reasoning', 'vision'],
+        contextWindow: 200_000,
+        maxOutput: 8_192,
+      },
+      {
+        modelId: 'gpt-5.5',
+        nickname: 'GPT-5.5',
+        capabilities: ['tool_use', 'reasoning', 'vision'],
+        contextWindow: 1_000_000,
+        maxOutput: 128_000,
+      },
+      {
+        modelId: 'gpt-5.4',
+        nickname: 'GPT-5.4',
+        capabilities: ['tool_use', 'reasoning', 'vision'],
+        contextWindow: 1_050_000,
+        maxOutput: 128_000,
+      },
+      {
+        modelId: 'gpt-5.4-mini',
+        nickname: 'GPT-5.4 Mini',
+        capabilities: ['tool_use', 'reasoning', 'vision'],
+        contextWindow: 400_000,
+        maxOutput: 128_000,
+      },
+      {
+        modelId: 'gemini-3.1-pro-preview',
+        nickname: 'Gemini 3.1 Pro',
+        capabilities: ['tool_use', 'reasoning', 'vision'],
+        contextWindow: 1_048_576,
+        maxOutput: 65_536,
+      },
+      {
+        modelId: 'deepseek-v4-pro',
+        nickname: 'DeepSeek V4 Pro',
+        capabilities: ['tool_use', 'reasoning'],
+        contextWindow: 163_840,
+        maxOutput: 163_840,
+      },
+      {
+        modelId: 'deepseek-v4-flash',
+        nickname: 'DeepSeek V4 Flash',
+        capabilities: ['tool_use'],
+        contextWindow: 128_000,
+        maxOutput: 32_768,
+      },
+      {
+        modelId: 'kimi-k2.6',
+        nickname: 'Kimi K2.6',
+        capabilities: ['tool_use'],
+        contextWindow: 131_072,
+        maxOutput: 32_768,
+      },
+      {
+        modelId: 'qwen3.6-plus',
+        nickname: 'Qwen 3.6 Plus',
+        capabilities: ['tool_use', 'reasoning'],
+        contextWindow: 128_000,
+        maxOutput: 32_768,
+      },
+      {
+        modelId: 'qwen3-coder-plus',
+        nickname: 'Qwen3 Coder Plus',
+        capabilities: ['tool_use', 'reasoning'],
+        contextWindow: 128_000,
+        maxOutput: 32_768,
+      },
+      {
+        modelId: 'glm-5.1',
+        nickname: 'GLM 5.1',
+        capabilities: ['tool_use', 'reasoning'],
+        contextWindow: 128_000,
+        maxOutput: 32_768,
+      },
+      {
+        modelId: 'MiniMax-M2.7',
+        nickname: 'MiniMax M2.7',
+        capabilities: ['tool_use'],
+        contextWindow: 1_048_576,
+        maxOutput: 65_536,
+      },
+      {
+        modelId: 'grok-4.20',
+        nickname: 'Grok 4.20',
+        capabilities: ['tool_use', 'reasoning', 'vision'],
+        contextWindow: 256_000,
+        maxOutput: 64_000,
+      },
+    ],
+  },
+  createModel: (config) => {
+    return new Kevoryn(
+      {
+        apiKey: config.effectiveApiKey,
+        model: config.model,
+        temperature: config.settings.temperature,
+        topP: config.settings.topP,
+        maxOutputTokens: config.settings.maxTokens,
+        stream: config.settings.stream,
+      },
+      config.dependencies
+    )
+  },
+  getDisplayName: (modelId, providerSettings) => {
+    return `Kevoryn API (${providerSettings?.models?.find((m) => m.modelId === modelId)?.nickname || modelId})`
+  },
+})
diff --git a/src/shared/providers/definitions/models/kevoryn.ts b/src/shared/providers/definitions/models/kevoryn.ts
new file mode 100644
index 0000000000..76048eb111
--- /dev/null
+++ b/src/shared/providers/definitions/models/kevoryn.ts
@@ -0,0 +1,41 @@
+import OpenAICompatible, { type OpenAICompatibleSettings } from '../../../models/openai-compatible'
+import type { ToolUseScope } from '../../../types'
+import type { ModelDependencies } from '../../../types/adapters'
+
+interface Options extends OpenAICompatibleSettings {}
+
+export default class Kevoryn extends OpenAICompatible {
+  public name = 'Kevoryn'
+  public options: Options
+  constructor(options: Omit<Options, 'apiHost'>, dependencies: ModelDependencies) {
+    const apiHost = 'https://api.kevoryn.com/v1'
+    super(
+      {
+        apiKey: options.apiKey,
+        apiHost,
+        model: options.model,
+        temperature: options.temperature,
+        topP: options.topP,
+        maxOutputTokens: options.maxOutputTokens,
+        stream: options.stream,
+      },
+      dependencies
+    )
+    this.options = {
+      ...options,
+      apiHost,
+    }
+  }
+
+  isSupportToolUse(_scope?: ToolUseScope) {
+    return true
+  }
+
+  isSupportVision() {
+    return true
+  }
+
+  isSupportReasoning() {
+    return true
+  }
+}
diff --git a/src/shared/providers/index.ts b/src/shared/providers/index.ts
index 06742110e7..b9497bf278 100644
--- a/src/shared/providers/index.ts
+++ b/src/shared/providers/index.ts
@@ -16,6 +16,7 @@ import './definitions/qwen-portal'
 import './definitions/minimax'
 import './definitions/moonshot'
 import './definitions/siliconflow'
+import './definitions/kevoryn'
 import './definitions/openrouter'
 import './definitions/ollama'
 import './definitions/lmstudio'
diff --git a/src/shared/types/provider.ts b/src/shared/types/provider.ts
index 8f12db4f03..ab01bb0538 100644
--- a/src/shared/types/provider.ts
+++ b/src/shared/types/provider.ts
@@ -24,7 +24,8 @@ export enum ModelProviderEnum {
   LMStudio = 'lm-studio',
   Perplexity = 'perplexity',
   XAI = 'xAI',
-  OpenRouter = 'openrouter',
+  OpenRouter = "openrouter",
+  Kevoryn = "kevoryn",
   Bedrock = 'bedrock',
   VercelAIGateway = 'vercel-ai-gateway',
   Custom = 'custom',