Add poe

src/main/presenter/configPresenter/providers.ts

@@ -199,6 +199,21 @@ export const DEFAULT_PROVIDERS: LLM_PROVIDER_BASE[] = [
       defaultBaseUrl: 'https://openrouter.ai/api/v1/'
     }
   },
+  {
+    id: 'poe',
+    name: 'Poe',
+    apiType: 'poe',
+    apiKey: '',
+    baseUrl: 'https://api.poe.com/v1',
+    enable: false,
+    websites: {
+      official: 'https://poe.com/',
+      apiKey: 'https://poe.com/api_key',
+      docs: 'https://creator.poe.com/docs/external-applications/openai-compatible-api',
+      models: 'https://api.poe.com/v1/models',
+      defaultBaseUrl: 'https://api.poe.com/v1'
+    }
+  },
   {
     id: '302ai',
     name: '302.AI',

src/main/presenter/llmProviderPresenter/index.ts

@@ -47,6 +47,7 @@ import { AihubmixProvider } from './providers/aihubmixProvider'
 import { _302AIProvider } from './providers/_302AIProvider'
 import { ModelscopeProvider } from './providers/modelscopeProvider'
 import { VercelAIGatewayProvider } from './providers/vercelAIGatewayProvider'
+import { PoeProvider } from './providers/poeProvider'
 
 // Rate limit configuration interface
 interface RateLimitConfig {
@@ -212,6 +213,8 @@ export class LLMProviderPresenter implements ILlmProviderPresenter {
           return new GroqProvider(provider, this.configPresenter)
         case 'vercel-ai-gateway':
           return new VercelAIGatewayProvider(provider, this.configPresenter)
+        case 'poe':
+          return new PoeProvider(provider, this.configPresenter)
         case 'aws-bedrock':
           return new AwsBedrockProvider(provider, this.configPresenter)
         default:
@@ -264,6 +267,8 @@ export class LLMProviderPresenter implements ILlmProviderPresenter {
           return new GrokProvider(provider, this.configPresenter)
         case 'vercel-ai-gateway':
           return new VercelAIGatewayProvider(provider, this.configPresenter)
+        case 'poe':
+          return new PoeProvider(provider, this.configPresenter)
         case 'aws-bedrock':
           return new AwsBedrockProvider(provider, this.configPresenter)
         default:

src/main/presenter/llmProviderPresenter/providers/poeProvider.ts

@@ -0,0 +1,25 @@
+import { LLM_PROVIDER, MODEL_META, IConfigPresenter } from '@shared/presenter'
+import { OpenAICompatibleProvider } from './openAICompatibleProvider'
+
+/**
+ * PoeProvider integrates Poe's OpenAI-compatible API surface with the shared
+ * BaseLLMProvider contract so the rest of the app can treat it just like
+ * any other OpenAI-style backend.
+ *
+ * Poe exposes hundreds of community and frontier models through a single
+ * endpoint. We reuse the OpenAICompatibleProvider implementation and only
+ * tweak metadata so the renderer can present a clearer group name.
+ */
+export class PoeProvider extends OpenAICompatibleProvider {
+  constructor(provider: LLM_PROVIDER, configPresenter: IConfigPresenter) {
+    super(provider, configPresenter)
+  }
+
+  protected async fetchOpenAIModels(options?: { timeout: number }): Promise<MODEL_META[]> {
+    const models = await super.fetchOpenAIModels(options)
+    return models.map((model) => ({
+      ...model,
+      group: 'Poe'
+    }))
+  }
+}

src/renderer/src/components/icons/ModelIcon.vue

@@ -44,6 +44,7 @@ import ppioColorIcon from '@/assets/llm-icons/ppio-color.svg?url'
 import tokenfluxColorIcon from '@/assets/llm-icons/tokenflux-color.svg?url'
 import moonshotColorIcon from '@/assets/llm-icons/moonshot.svg?url'
 import openrouterColorIcon from '@/assets/llm-icons/openrouter.svg?url'
+import poeColorIcon from '@/assets/llm-icons/poe-color.svg?url'
 import geminiColorIcon from '@/assets/llm-icons/gemini-color.svg?url'
 import githubColorIcon from '@/assets/llm-icons/github.svg?url'
 import azureOpenaiColorIcon from '@/assets/llm-icons/azure-color.svg?url'
@@ -119,6 +120,7 @@ const icons = {
   qwen: qwenColorIcon,
   moonshot: moonshotColorIcon,
   openrouter: openrouterColorIcon,
+  poe: poeColorIcon,
   gemini: geminiColorIcon,
   github: githubColorIcon,
   anthropic: claudeColorIcon,