@lobehub/chat 0.146.1 → 0.147.0

package/src/libs/agent-runtime/moonshot/index.ts

@@ -1,78 +1,15 @@
-import { OpenAIStream, StreamingTextResponse } from 'ai';
-import OpenAI, { ClientOptions } from 'openai';
-
-import { LobeRuntimeAI } from '../BaseAI';
 import { AgentRuntimeErrorType } from '../error';
-import { ChatCompetitionOptions, ChatStreamPayload, ModelProvider } from '../types';
-import { AgentRuntimeError } from '../utils/createError';
-import { debugStream } from '../utils/debugStream';
-import { desensitizeUrl } from '../utils/desensitizeUrl';
-import { handleOpenAIError } from '../utils/handleOpenAIError';
-
-const DEFAULT_BASE_URL = 'https://api.moonshot.cn/v1';
-
-export class LobeMoonshotAI implements LobeRuntimeAI {
-  private client: OpenAI;
-
-  baseURL: string;
-
-  constructor({ apiKey, baseURL = DEFAULT_BASE_URL, ...res }: ClientOptions) {
-    if (!apiKey) throw AgentRuntimeError.createError(AgentRuntimeErrorType.InvalidMoonshotAPIKey);
-
-    this.client = new OpenAI({ apiKey, baseURL, ...res });
-    this.baseURL = this.client.baseURL;
-  }
-
-  async chat(payload: ChatStreamPayload, options?: ChatCompetitionOptions) {
-    try {
-      const response = await this.client.chat.completions.create(
-        payload as unknown as OpenAI.ChatCompletionCreateParamsStreaming,
-      );
-      const [prod, debug] = response.tee();
-
-      if (process.env.DEBUG_MOONSHOT_CHAT_COMPLETION === '1') {
-        debugStream(debug.toReadableStream()).catch(console.error);
-      }
-
-      return new StreamingTextResponse(OpenAIStream(prod, options?.callback), {
-        headers: options?.headers,
-      });
-    } catch (error) {
-      let desensitizedEndpoint = this.baseURL;
-
-      if (this.baseURL !== DEFAULT_BASE_URL) {
-        desensitizedEndpoint = desensitizeUrl(this.baseURL);
-      }
-
-      if ('status' in (error as any)) {
-        switch ((error as Response).status) {
-          case 401: {
-            throw AgentRuntimeError.chat({
-              endpoint: desensitizedEndpoint,
-              error: error as any,
-              errorType: AgentRuntimeErrorType.InvalidMoonshotAPIKey,
-              provider: ModelProvider.Moonshot,
-            });
-          }
-
-          default: {
-            break;
-          }
-        }
-      }
-
-      const { errorResult, RuntimeError } = handleOpenAIError(error);
-
-      const errorType = RuntimeError || AgentRuntimeErrorType.MoonshotBizError;
-
-      throw AgentRuntimeError.chat({
-        endpoint: desensitizedEndpoint,
-        error: errorResult,
-        errorType,
-        provider: ModelProvider.Moonshot,
-      });
-    }
-  }
-}
-
-export default LobeMoonshotAI;
+import { ModelProvider } from '../types';
+import { LobeOpenAICompatibleFactory } from '../utils/openaiCompatibleFactory';
+
+export const LobeMoonshotAI = LobeOpenAICompatibleFactory({
+  baseURL: 'https://api.moonshot.cn/v1',
+  debug: {
+    chatCompletion: () => process.env.DEBUG_MOONSHOT_CHAT_COMPLETION === '1',
+  },
+  errorType: {
+    bizError: AgentRuntimeErrorType.MoonshotBizError,
+    invalidAPIKey: AgentRuntimeErrorType.InvalidMoonshotAPIKey,
+  },
+  provider: ModelProvider.Moonshot,
+});

package/src/libs/agent-runtime/openai/__snapshots__/index.test.ts.snap

@@ -0,0 +1,99 @@
+// Vitest Snapshot v1, https://vitest.dev/guide/snapshot.html
+
+exports[`LobeOpenAI > models > should get models 1`] = `
+[
+  {
+    "description": "GPT 3.5 Turbo，适用于各种文本生成和理解任务",
+    "displayName": "GPT-3.5 Turbo",
+    "enabled": true,
+    "functionCall": true,
+    "id": "gpt-3.5-turbo",
+    "tokens": 16385,
+  },
+  {
+    "displayName": "GPT-3.5 Turbo 16K",
+    "id": "gpt-3.5-turbo-16k",
+    "tokens": 16385,
+  },
+  {
+    "displayName": "GPT-3.5 Turbo 16K (0613)",
+    "id": "gpt-3.5-turbo-16k-0613",
+    "legacy": true,
+    "tokens": 4096,
+  },
+  {
+    "displayName": "GPT-4 Turbo Vision Preview (1106)",
+    "id": "gpt-4-1106-vision-preview",
+    "tokens": 128000,
+    "vision": true,
+  },
+  {
+    "id": "gpt-3.5-turbo-instruct-0914",
+  },
+  {
+    "displayName": "GPT-4 Turbo Preview (0125)",
+    "functionCall": true,
+    "id": "gpt-4-0125-preview",
+    "tokens": 128000,
+  },
+  {
+    "displayName": "GPT-4 Turbo Preview",
+    "enabled": true,
+    "functionCall": true,
+    "id": "gpt-4-turbo-preview",
+    "tokens": 128000,
+  },
+  {
+    "displayName": "GPT-3.5 Turbo Instruct",
+    "id": "gpt-3.5-turbo-instruct",
+    "tokens": 4096,
+  },
+  {
+    "id": "gpt-3.5-turbo-0301",
+  },
+  {
+    "displayName": "GPT-3.5 Turbo (0613)",
+    "id": "gpt-3.5-turbo-0613",
+    "legacy": true,
+    "tokens": 4096,
+  },
+  {
+    "displayName": "GPT-3.5 Turbo (1106)",
+    "functionCall": true,
+    "id": "gpt-3.5-turbo-1106",
+    "tokens": 16385,
+  },
+  {
+    "displayName": "GPT-4 Turbo Preview (1106)",
+    "functionCall": true,
+    "id": "gpt-4-1106-preview",
+    "tokens": 128000,
+  },
+  {
+    "description": "GPT-4 视觉预览版，支持视觉任务",
+    "displayName": "GPT-4 Turbo Vision Preview",
+    "enabled": true,
+    "id": "gpt-4-vision-preview",
+    "tokens": 128000,
+    "vision": true,
+  },
+  {
+    "displayName": "GPT-4",
+    "functionCall": true,
+    "id": "gpt-4",
+    "tokens": 8192,
+  },
+  {
+    "displayName": "GPT-3.5 Turbo (0125)",
+    "functionCall": true,
+    "id": "gpt-3.5-turbo-0125",
+    "tokens": 16385,
+  },
+  {
+    "displayName": "GPT-4 (0613)",
+    "functionCall": true,
+    "id": "gpt-4-0613",
+    "tokens": 8192,
+  },
+]
+`;

package/src/libs/agent-runtime/openai/fixtures/openai-models.json

@@ -0,0 +1,170 @@
+[
+  {
+    "id": "whisper-1",
+    "object": "model",
+    "created": 1677532384,
+    "owned_by": "openai-internal"
+  },
+  {
+    "id": "davinci-002",
+    "object": "model",
+    "created": 1692634301,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-3.5-turbo",
+    "object": "model",
+    "created": 1677610602,
+    "owned_by": "openai"
+  },
+  {
+    "id": "dall-e-2",
+    "object": "model",
+    "created": 1698798177,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-3.5-turbo-16k",
+    "object": "model",
+    "created": 1683758102,
+    "owned_by": "openai-internal"
+  },
+  {
+    "id": "tts-1-hd-1106",
+    "object": "model",
+    "created": 1699053533,
+    "owned_by": "system"
+  },
+  {
+    "id": "tts-1-hd",
+    "object": "model",
+    "created": 1699046015,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-3.5-turbo-16k-0613",
+    "object": "model",
+    "created": 1685474247,
+    "owned_by": "openai"
+  },
+  {
+    "id": "text-embedding-3-large",
+    "object": "model",
+    "created": 1705953180,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-4-1106-vision-preview",
+    "object": "model",
+    "created": 1711473033,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-3.5-turbo-instruct-0914",
+    "object": "model",
+    "created": 1694122472,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-4-0125-preview",
+    "object": "model",
+    "created": 1706037612,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-4-turbo-preview",
+    "object": "model",
+    "created": 1706037777,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-3.5-turbo-instruct",
+    "object": "model",
+    "created": 1692901427,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-3.5-turbo-0301",
+    "object": "model",
+    "created": 1677649963,
+    "owned_by": "openai"
+  },
+  {
+    "id": "gpt-3.5-turbo-0613",
+    "object": "model",
+    "created": 1686587434,
+    "owned_by": "openai"
+  },
+  {
+    "id": "tts-1",
+    "object": "model",
+    "created": 1681940951,
+    "owned_by": "openai-internal"
+  },
+  {
+    "id": "dall-e-3",
+    "object": "model",
+    "created": 1698785189,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-3.5-turbo-1106",
+    "object": "model",
+    "created": 1698959748,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-4-1106-preview",
+    "object": "model",
+    "created": 1698957206,
+    "owned_by": "system"
+  },
+  {
+    "id": "babbage-002",
+    "object": "model",
+    "created": 1692634615,
+    "owned_by": "system"
+  },
+  {
+    "id": "tts-1-1106",
+    "object": "model",
+    "created": 1699053241,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-4-vision-preview",
+    "object": "model",
+    "created": 1698894917,
+    "owned_by": "system"
+  },
+  {
+    "id": "text-embedding-3-small",
+    "object": "model",
+    "created": 1705948997,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-4",
+    "object": "model",
+    "created": 1687882411,
+    "owned_by": "openai"
+  },
+  {
+    "id": "text-embedding-ada-002",
+    "object": "model",
+    "created": 1671217299,
+    "owned_by": "openai-internal"
+  },
+  {
+    "id": "gpt-3.5-turbo-0125",
+    "object": "model",
+    "created": 1706048358,
+    "owned_by": "system"
+  },
+  {
+    "id": "gpt-4-0613",
+    "object": "model",
+    "created": 1686588896,
+    "owned_by": "openai"
+  }
+]

package/src/libs/agent-runtime/openai/index.test.ts

@@ -3,16 +3,17 @@ import OpenAI from 'openai';
 import { Mock, afterEach, beforeEach, describe, expect, it, vi } from 'vitest';
 
 // 引入模块以便于对函数进行spy
-import { ChatStreamCallbacks } from '@/libs/agent-runtime';
+import { ChatStreamCallbacks, LobeOpenAICompatibleRuntime } from '@/libs/agent-runtime';
 
 import * as debugStreamModule from '../utils/debugStream';
+import officalOpenAIModels from './fixtures/openai-models.json';
 import { LobeOpenAI } from './index';
 
 // Mock the console.error to avoid polluting test output
 vi.spyOn(console, 'error').mockImplementation(() => {});
 
 describe('LobeOpenAI', () => {
-  let instance: LobeOpenAI;
+  let instance: LobeOpenAICompatibleRuntime;
 
   beforeEach(() => {
     instance = new LobeOpenAI({ apiKey: 'test' });
@@ -21,60 +22,13 @@ describe('LobeOpenAI', () => {
     vi.spyOn(instance['client'].chat.completions, 'create').mockResolvedValue(
       new ReadableStream() as any,
     );
+    vi.spyOn(instance['client'].models, 'list').mockResolvedValue({ data: [] } as any);
   });
 
   afterEach(() => {
     vi.clearAllMocks();
   });
 
-  describe('init', () => {
-    it('should correctly initialize with Azure options', () => {
-      const baseURL = 'https://abc.com';
-      const modelName = 'abc';
-      const client = new LobeOpenAI({
-        apiKey: 'test',
-        useAzure: true,
-        baseURL,
-        azureOptions: {
-          apiVersion: '2023-08-01-preview',
-          model: 'abc',
-        },
-      });
-
-      expect(client.baseURL).toEqual(baseURL + '/openai/deployments/' + modelName);
-    });
-
-    describe('initWithAzureOpenAI', () => {
-      it('should correctly initialize with Azure options', () => {
-        const baseURL = 'https://abc.com';
-        const modelName = 'abc';
-        const client = LobeOpenAI.initWithAzureOpenAI({
-          apiKey: 'test',
-          useAzure: true,
-          baseURL,
-          azureOptions: {
-            apiVersion: '2023-08-01-preview',
-            model: 'abc',
-          },
-        });
-
-        expect(client.baseURL).toEqual(baseURL + '/openai/deployments/' + modelName);
-      });
-
-      it('should use default Azure options when not explicitly provided', () => {
-        const baseURL = 'https://abc.com';
-
-        const client = LobeOpenAI.initWithAzureOpenAI({
-          apiKey: 'test',
-          useAzure: true,
-          baseURL,
-        });
-
-        expect(client.baseURL).toEqual(baseURL + '/openai/deployments/');
-      });
-    });
-  });
-
   describe('chat', () => {
     it('should return a StreamingTextResponse on successful API call', async () => {
       // Arrange
@@ -330,4 +284,15 @@ describe('LobeOpenAI', () => {
       });
     });
   });
+
+  describe('models', () => {
+    it('should get models', async () => {
+      // mock the models.list method
+      (instance['client'].models.list as Mock).mockResolvedValue({ data: officalOpenAIModels });
+
+      const list = await instance.models();
+
+      expect(list).toMatchSnapshot();
+    });
+  });
 });

package/src/libs/agent-runtime/openai/index.ts

@@ -1,108 +1,16 @@
-import { OpenAIStream, StreamingTextResponse } from 'ai';
-import OpenAI, { ClientOptions } from 'openai';
-import urlJoin from 'url-join';
-
-import { ChatStreamPayload } from '@/types/openai/chat';
-
-import { LobeRuntimeAI } from '../BaseAI';
 import { AgentRuntimeErrorType } from '../error';
-import { ChatCompetitionOptions, ModelProvider } from '../types';
-import { AgentRuntimeError } from '../utils/createError';
-import { debugStream } from '../utils/debugStream';
-import { desensitizeUrl } from '../utils/desensitizeUrl';
-import { handleOpenAIError } from '../utils/handleOpenAIError';
-
-const DEFAULT_BASE_URL = 'https://api.openai.com/v1';
-
-interface LobeOpenAIOptions extends ClientOptions {
-  azureOptions?: {
-    apiVersion?: string;
-    model?: string;
-  };
-  useAzure?: boolean;
-}
-
-export class LobeOpenAI implements LobeRuntimeAI {
-  private client: OpenAI;
-
-  constructor(options: LobeOpenAIOptions) {
-    if (!options.apiKey) throw AgentRuntimeError.createError(AgentRuntimeErrorType.NoOpenAIAPIKey);
-
-    if (options.useAzure) {
-      this.client = LobeOpenAI.initWithAzureOpenAI(options);
-    } else {
-      this.client = new OpenAI(options);
-    }
-
-    this.baseURL = this.client.baseURL;
-  }
-
-  baseURL: string;
-
-  async chat(payload: ChatStreamPayload, options?: ChatCompetitionOptions) {
-    // ============  1. preprocess messages   ============ //
-    const { messages, ...params } = payload;
-
-    // ============  2. send api   ============ //
-
-    try {
-      const response = await this.client.chat.completions.create(
-        {
-          messages,
-          ...params,
-          stream: true,
-        } as unknown as OpenAI.ChatCompletionCreateParamsStreaming,
-        { headers: { Accept: '*/*' } },
-      );
-
-      const [prod, debug] = response.tee();
-
-      if (process.env.DEBUG_OPENAI_CHAT_COMPLETION === '1') {
-        debugStream(debug.toReadableStream()).catch(console.error);
-      }
-
-      return new StreamingTextResponse(OpenAIStream(prod, options?.callback), {
-        headers: options?.headers,
-      });
-    } catch (error) {
-      const { errorResult, RuntimeError } = handleOpenAIError(error);
-
-      const errorType = RuntimeError || AgentRuntimeErrorType.OpenAIBizError;
-
-      let desensitizedEndpoint = this.baseURL;
-
-      // refs: https://github.com/lobehub/lobe-chat/issues/842
-      if (this.baseURL !== DEFAULT_BASE_URL) {
-        desensitizedEndpoint = desensitizeUrl(this.baseURL);
-      }
-
-      throw AgentRuntimeError.chat({
-        endpoint: desensitizedEndpoint,
-        error: errorResult,
-        errorType,
-        provider: ModelProvider.OpenAI,
-      });
-    }
-  }
-
-  static initWithAzureOpenAI(options: LobeOpenAIOptions) {
-    const endpoint = options.baseURL!;
-    const model = options.azureOptions?.model || '';
-
-    // refs: https://test-001.openai.azure.com/openai/deployments/gpt-35-turbo
-    const baseURL = urlJoin(endpoint, `/openai/deployments/${model.replace('.', '')}`);
-
-    const apiVersion = options.azureOptions?.apiVersion || '2023-08-01-preview';
-    const apiKey = options.apiKey!;
-
-    const config: ClientOptions = {
-      ...options,
-      apiKey,
-      baseURL,
-      defaultHeaders: { 'api-key': apiKey },
-      defaultQuery: { 'api-version': apiVersion },
-    };
-
-    return new OpenAI(config);
-  }
-}
+import { ModelProvider } from '../types';
+import { LobeOpenAICompatibleFactory } from '../utils/openaiCompatibleFactory';
+
+export const LobeOpenAI = LobeOpenAICompatibleFactory({
+  baseURL: 'https://api.openai.com/v1',
+  debug: {
+    chatCompletion: () => process.env.DEBUG_OPENAI_CHAT_COMPLETION === '1',
+  },
+  errorType: {
+    bizError: AgentRuntimeErrorType.OpenAIBizError,
+    invalidAPIKey: AgentRuntimeErrorType.NoOpenAIAPIKey,
+  },
+
+  provider: ModelProvider.OpenAI,
+});

package/src/libs/agent-runtime/openrouter/__snapshots__/index.test.ts.snap

@@ -0,0 +1,82 @@
+// Vitest Snapshot v1, https://vitest.dev/guide/snapshot.html
+
+exports[`LobeOpenRouterAI > models > should get models 1`] = `
+[
+  {
+    "description": "LLaVA is a large multimodal model that combines a vision encoder and Vicuna for general-purpose visual and language understanding, achieving impressive chat capabilities mimicking [GPT-4](/models/openai/gpt-4-vision-preview) and setting a new state-of-the-art accuracy on Science QA
+
+#multimodal",
+    "displayName": "Llava 13B",
+    "enabled": false,
+    "functionCall": false,
+    "id": "haotian-liu/llava-13b",
+    "maxTokens": undefined,
+    "tokens": 2048,
+    "vision": true,
+  },
+  {
+    "description": "This vision-language model builds on innovations from the popular [OpenHermes-2.5](/models/teknium/openhermes-2.5-mistral-7b) model, by Teknium. It adds vision support, and is trained on a custom dataset enriched with function calling
+
+This project is led by [qnguyen3](https://twitter.com/stablequan) and [teknium](https://twitter.com/Teknium1).
+
+#multimodal",
+    "displayName": "Nous: Hermes 2 Vision 7B (alpha)",
+    "enabled": false,
+    "functionCall": true,
+    "id": "nousresearch/nous-hermes-2-vision-7b",
+    "maxTokens": undefined,
+    "tokens": 4096,
+    "vision": true,
+  },
+  {
+    "description": "GPT-3.5 Turbo is OpenAI's fastest model. It can understand and generate natural language or code, and is optimized for chat and traditional completion tasks.
+
+Updated by OpenAI to point to the [latest version of GPT-3.5](/models?q=openai/gpt-3.5). Training data up to Sep 2021.",
+    "displayName": "OpenAI: GPT-3.5 Turbo",
+    "enabled": true,
+    "functionCall": false,
+    "id": "openai/gpt-3.5-turbo",
+    "maxTokens": 4096,
+    "tokens": 16385,
+    "vision": false,
+  },
+  {
+    "description": "Ability to understand images, in addition to all other [GPT-4 Turbo capabilties](/models/openai/gpt-4-turbo). Training data: up to Apr 2023.
+
+**Note:** heavily rate limited by OpenAI while in preview.
+
+#multimodal",
+    "displayName": "OpenAI: GPT-4 Vision",
+    "enabled": true,
+    "functionCall": false,
+    "id": "openai/gpt-4-vision-preview",
+    "maxTokens": 4096,
+    "tokens": 128000,
+    "vision": true,
+  },
+  {
+    "description": "Gemma by Google is an advanced, open-source language model family, leveraging the latest in decoder-only, text-to-text technology. It offers English language capabilities across text generation tasks like question answering, summarization, and reasoning. The Gemma 7B variant is comparable in performance to leading open source models.
+
+Usage of Gemma is subject to Google's [Gemma Terms of Use](https://ai.google.dev/gemma/terms).",
+    "displayName": "Google: Gemma 7B",
+    "enabled": true,
+    "functionCall": false,
+    "id": "google/gemma-7b-it",
+    "maxTokens": undefined,
+    "tokens": 8192,
+    "vision": false,
+  },
+  {
+    "description": "One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay. #merge
+
+Note: this is a higher-throughput version of [this model](/models/gryphe/mythomax-l2-13b), and may have higher prices and slightly different outputs.",
+    "displayName": "MythoMax 13B (nitro)",
+    "enabled": false,
+    "functionCall": false,
+    "id": "gryphe/mythomax-l2-13b:nitro",
+    "maxTokens": undefined,
+    "tokens": 4096,
+    "vision": false,
+  },
+]
+`;