🐳 feat: RAG for Default Docker Compose Files + Docs Update (danny-avi…

…la#2246)

* refactor(deploy-compose.yml): use long-syntax to avoid implicit folder creation of librechat.yaml

* refactor(docker-compose.override.yml.example): use long-syntax to avoid implicit folder creation of librechat.yaml

* chore: add simple health check for RAG_API_URL

* chore: improve axios error handling, adding `logAxiosError`

* chore: more informative message detailing RAG_API_URL path

* feat: add rag_api and vectordb to default compose file

* chore(rag.yml): update standalone rag compose file to use RAG_PORT

* chore: documentation updates

* docs: Update rag_api.md with images

* Update rag_api.md

* Update rag_api.md, assistants clarification

* add RAG API note to breaking changes

README.md

@@ -42,9 +42,11 @@
 
 - 🖥️ UI matching ChatGPT, including Dark mode, Streaming, and latest updates
 - 💬 Multimodal Chat:
-  - Upload and analyze images with GPT-4 and Gemini Vision 📸
-  - General file support now available through the Assistants API integration. 🗃️
-  - Local RAG in Active Development 🚧
+    - Upload and analyze images with Claude 3, GPT-4, and Gemini Vision 📸
+    - Chat with Files using Custom Endpoints, OpenAI, Azure, Anthropic, & Google. 🗃️
+    - Advanced Agents with Files, Code Interpreter, Tools, and API Actions 🔦
+      - Available through the [OpenAI Assistants API](https://platform.openai.com/docs/assistants/overview) 🌤️
+      - Non-OpenAI Agents in Active Development 🚧
 - 🌎 Multilingual UI:
   - English, 中文, Deutsch, Español, Français, Italiano, Polski, Português Brasileiro,
   - Русский, 日本語, Svenska, 한국어, Tiếng Việt, 繁體中文, العربية, Türkçe, Nederlands, עברית
@@ -55,7 +57,9 @@
 - 🔍 Search all messages/conversations
 - 🔌 Plugins, including web access, image generation with DALL-E-3 and more
 - 👥 Multi-User, Secure Authentication with Moderation and Token spend tools
-- ⚙️ Configure Proxy, Reverse Proxy, Docker, many Deployment options, and completely Open-Source
+- ⚙️ Configure Proxy, Reverse Proxy, Docker, & many Deployment options
+- 📖 Completely Open-Source & Built in Public
+- 🧑‍🤝‍🧑 Community-driven development, support, and feedback
 
 [For a thorough review of our features, see our docs here](https://docs.librechat.ai/features/plugins/introduction.html) 📚
 

api/server/services/AppService.js

@@ -156,6 +156,17 @@ const AppService = async (app) => {
     };
   }
 
+  try {
+    const response = await fetch(`${process.env.RAG_API_URL}/health`);
+    if (response?.ok && response?.status === 200) {
+      logger.info(`RAG API is running and reachable at ${process.env.RAG_API_URL}.`);
+    }
+  } catch (error) {
+    logger.warn(
+      `RAG API is either not running or not reachable at ${process.env.RAG_API_URL}, you may experience errors with file uploads.`,
+    );
+  }
+
   app.locals = {
     socialLogins,
     availableTools,

api/server/services/ModelService.js

@@ -1,11 +1,8 @@
 const axios = require('axios');
 const { HttpsProxyAgent } = require('https-proxy-agent');
 const { EModelEndpoint, defaultModels, CacheKeys } = require('librechat-data-provider');
-const { extractBaseURL, inputSchema, processModelData } = require('~/utils');
+const { extractBaseURL, inputSchema, processModelData, logAxiosError } = require('~/utils');
 const getLogStores = require('~/cache/getLogStores');
-const { logger } = require('~/config');
-
-// const { getAzureCredentials, genAzureChatCompletion } = require('~/utils/');
 
 const { openAIApiKey, userProvidedOpenAI } = require('./Config/EndpointService').config;
 
@@ -77,29 +74,7 @@ const fetchModels = async ({
     models = input.data.map((item) => item.id);
   } catch (error) {
     const logMessage = `Failed to fetch models from ${azure ? 'Azure ' : ''}${name} API`;
-    if (error.response) {
-      logger.error(
-        `${logMessage} The request was made and the server responded with a status code that falls out of the range of 2xx: ${
-          error.message ? error.message : ''
-        }`,
-        {
-          headers: error.response.headers,
-          status: error.response.status,
-          data: error.response.data,
-        },
-      );
-    } else if (error.request) {
-      logger.error(
-        `${logMessage} The request was made but no response was received: ${
-          error.message ? error.message : ''
-        }`,
-        {
-          request: error.request,
-        },
-      );
-    } else {
-      logger.error(`${logMessage} Something happened in setting up the request`, error);
-    }
+    logAxiosError({ message: logMessage, error });
   }
 
   return models;

api/server/services/Runs/methods.js

@@ -1,6 +1,6 @@
 const axios = require('axios');
 const { EModelEndpoint } = require('librechat-data-provider');
-const { logger } = require('~/config');
+const { logAxiosError } = require('~/utils');
 
 /**
  * @typedef {Object} RetrieveOptions
@@ -54,33 +54,8 @@ async function retrieveRun({ thread_id, run_id, timeout, openai }) {
     const response = await axios.get(url, axiosConfig);
     return response.data;
   } catch (error) {
-    const logMessage = '[retrieveRun] Failed to retrieve run data:';
-    const timedOutMessage = 'Cannot read properties of undefined (reading \'status\')';
-    if (error?.response && error?.response?.status) {
-      logger.error(
-        `${logMessage} The request was made and the server responded with a status code that falls out of the range of 2xx: ${
-          error.message ? error.message : ''
-        }`,
-        {
-          headers: error.response.headers,
-          status: error.response.status,
-          data: error.response.data,
-        },
-      );
-    } else if (error.request) {
-      logger.error(
-        `${logMessage} The request was made but no response was received: ${
-          error.message ? error.message : ''
-        }`,
-        {
-          request: error.request,
-        },
-      );
-    } else if (error?.message && !error?.message?.includes(timedOutMessage)) {
-      logger.error(`${logMessage} Something happened in setting up the request`, {
-        message: error.message,
-      });
-    }
+    const message = '[retrieveRun] Failed to retrieve run data:';
+    logAxiosError({ message, error });
     throw error;
   }
 }

api/utils/index.js

@@ -1,13 +1,15 @@
 const loadYaml = require('./loadYaml');
 const tokenHelpers = require('./tokens');
 const azureUtils = require('./azureUtils');
+const logAxiosError = require('./logAxiosError');
 const extractBaseURL = require('./extractBaseURL');
 const findMessageContent = require('./findMessageContent');
 
 module.exports = {
-  ...azureUtils,
+  loadYaml,
   ...tokenHelpers,
+  ...azureUtils,
+  logAxiosError,
   extractBaseURL,
   findMessageContent,
-  loadYaml,
 };

api/utils/logAxiosError.js

@@ -0,0 +1,45 @@
+const { logger } = require('~/config');
+
+/**
+ * Logs Axios errors based on the error object and a custom message.
+ *
+ * @param {Object} options - The options object.
+ * @param {string} options.message - The custom message to be logged.
+ * @param {Error} options.error - The Axios error object.
+ */
+const logAxiosError = ({ message, error }) => {
+  const timedOutMessage = 'Cannot read properties of undefined (reading \'status\')';
+  if (error.response) {
+    logger.error(
+      `${message} The request was made and the server responded with a status code that falls out of the range of 2xx: ${
+        error.message ? error.message : ''
+      }. Error response data:\n`,
+      {
+        headers: error.response?.headers,
+        status: error.response?.status,
+        data: error.response?.data,
+      },
+    );
+  } else if (error.request) {
+    logger.error(
+      `${message} The request was made but no response was received: ${
+        error.message ? error.message : ''
+      }. Error Request:\n`,
+      {
+        request: error.request,
+      },
+    );
+  } else if (error?.message?.includes(timedOutMessage)) {
+    logger.error(
+      `${message}\nThe request either timed out or was unsuccessful. Error message:\n`,
+      error,
+    );
+  } else {
+    logger.error(
+      `${message}\nSomething happened in setting up the request. Error message:\n`,
+      error,
+    );
+  }
+};
+
+module.exports = logAxiosError;

deploy-compose.yml

@@ -11,6 +11,7 @@ services:
       - 3080:3080
     depends_on:
       - mongodb
+      - rag_api
     restart: always
     extra_hosts:
     - "host.docker.internal:host-gateway"
@@ -21,11 +22,14 @@ services:
       - NODE_ENV=production
       - MONGO_URI=mongodb://mongodb:27017/LibreChat
       - MEILI_HOST=http://meilisearch:7700
+      - RAG_PORT=${RAG_PORT:-8000}
+      - RAG_API_URL=http://rag_api:${RAG_PORT:-8000}
     volumes:
+      - type: bind
+        source: ./librechat.yaml
+        target: /app/librechat.yaml
       - ./images:/app/client/public/images
-      - ./librechat.yaml:/app/librechat.yaml
       - ./logs:/app/api/logs
-      - ./uploads:/app/uploads
   client:
     build:
       context: .
@@ -61,3 +65,25 @@ services:
       - MEILI_NO_ANALYTICS=true
     volumes:
       - ./meili_data_v1.7:/meili_data
+  vectordb:
+    image: ankane/pgvector:latest
+    environment:
+      POSTGRES_DB: mydatabase
+      POSTGRES_USER: myuser
+      POSTGRES_PASSWORD: mypassword
+    restart: always
+    volumes:
+      - pgdata2:/var/lib/postgresql/data
+  rag_api:
+    image: ghcr.io/danny-avila/librechat-rag-api-dev-lite:latest
+    environment:
+      - DB_HOST=vectordb
+      - RAG_PORT=${RAG_PORT:-8000}
+    restart: always
+    depends_on:
+      - vectordb
+    env_file:
+      - .env
+
+volumes:
+  pgdata2:

docker-compose.override.yml.example

@@ -16,7 +16,9 @@ version: '3.4'
 # services:
 #   api:
 #     volumes:
-#       - ./librechat.yaml:/app/librechat.yaml
+#     - type: bind
+#       source: ./librechat.yaml
+#       target: /app/librechat.yaml
 #     image: ghcr.io/danny-avila/librechat:latest
 
 # ---------------------------------------------------
@@ -26,7 +28,9 @@ version: '3.4'
 # # USE LIBRECHAT CONFIG FILE
 #   api:
 #     volumes:
-#       - ./librechat.yaml:/app/librechat.yaml
+#     - type: bind
+#       source: ./librechat.yaml
+#       target: /app/librechat.yaml
 
 # # LOCAL BUILD
 #   api:
@@ -93,6 +97,10 @@ version: '3.4'
 #     ports:
 #       - 7700:7700
 
+# # USE RAG API IMAGE WITH LOCAL EMBEDDINGS SUPPORT
+#  rag_api:
+#    image: ghcr.io/danny-avila/librechat-rag-api-dev:latest
+
 # # ADD OLLAMA
 #  ollama:
 #    image: ollama/ollama:latest

docker-compose.yml

@@ -10,6 +10,7 @@ services:
       - "${PORT}:${PORT}"
     depends_on:
       - mongodb
+      - rag_api
     image: ghcr.io/danny-avila/librechat-dev:latest
     restart: always
     user: "${UID}:${GID}"
@@ -19,10 +20,13 @@ services:
       - HOST=0.0.0.0
       - MONGO_URI=mongodb://mongodb:27017/LibreChat
       - MEILI_HOST=http://meilisearch:7700
+      - RAG_PORT=${RAG_PORT:-8000}
+      - RAG_API_URL=http://rag_api:${RAG_PORT:-8000}
     volumes:
-      - ./.env:/app/.env
+      - type: bind
+        source: ./.env
+        target: /app/.env
       - ./images:/app/client/public/images
-      - ./uploads:/app/uploads
       - ./logs:/app/api/logs
   mongodb:
     container_name: chat-mongodb
@@ -42,3 +46,25 @@ services:
       - MEILI_NO_ANALYTICS=true
     volumes:
       - ./meili_data_v1.7:/meili_data
+  vectordb:
+    image: ankane/pgvector:latest
+    environment:
+      POSTGRES_DB: mydatabase
+      POSTGRES_USER: myuser
+      POSTGRES_PASSWORD: mypassword
+    restart: always
+    volumes:
+      - pgdata2:/var/lib/postgresql/data
+  rag_api:
+    image: ghcr.io/danny-avila/librechat-rag-api-dev-lite:latest
+    environment:
+      - DB_HOST=vectordb
+      - RAG_PORT=${RAG_PORT:-8000}
+    restart: always
+    depends_on:
+      - vectordb
+    env_file:
+      - .env
+
+volumes:
+  pgdata2:

docs/features/index.md

@@ -9,6 +9,8 @@ weight: 2
 ---
 
 * 🤖[Custom Endpoints](../install/configuration/custom_config.md)
+* 🗃️ [RAG API (Chat with Files)](./rag_api.md)
+* 🔖 [Presets](./presets.md)
 * 🔌[Plugins](./plugins/index.md)
     * 🔌 [Introduction](./plugins/introduction.md)
     * 🛠️ [Make Your Own](./plugins/make_your_own.md)
@@ -17,7 +19,6 @@ weight: 2
     * 🖌️ [Stable Diffusion](./plugins/stable_diffusion.md)
     * 🧠 [Wolfram|Alpha](./plugins/wolfram.md)
     * ⚡ [Azure AI Search](./plugins/azure_ai_search.md)
-* 🔖 [Presets](./presets.md)
 
 ---
 

docs/features/mod_system.md

@@ -1,7 +1,7 @@
 ---
 title: 🔨 Automated Moderation
 description: The Automated Moderation System uses a scoring mechanism to track user violations. As users commit actions like excessive logins, registrations, or messaging, they accumulate violation scores. Upon reaching a set threshold, the user and their IP are temporarily banned. This system ensures platform security by monitoring and penalizing rapid or suspicious activities.
-weight: -8
+weight: -7
 ---
 ## Automated Moderation System (optional)
 The Automated Moderation System uses a scoring mechanism to track user violations. As users commit actions like excessive logins, registrations, or messaging, they accumulate violation scores. Upon reaching a set threshold, the user and their IP are temporarily banned. This system ensures platform security by monitoring and penalizing rapid or suspicious activities.

docs/features/plugins/index.md

@@ -1,7 +1,7 @@
 ---
 title: Plugins
 description: 🔌 All about plugins, how to make them, how use the official ChatGPT plugins, and how to configure custom plugins. 
-weight: -10
+weight: -9
 ---
 
 # Plugins