### Azure OpenAI Configuration Example

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/model-providers.md

Example environment variable setup for Azure OpenAI. Ensure to replace placeholders with your specific Azure details.

```bash
AZURE_URL=https://{resource-name}.openai.azure.com/openai/deployments/{deployment-name}
AZURE_API_KEY=your_azure_key
AZURE_API_VERSION=2022-06-01-preview
```

--------------------------------

### Docker Quick Start

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/quick-start-reference.md

Quickly run the NextChat Docker image locally. Ensure you have Docker installed and replace placeholder API keys.

```bash
# Pull the image
docker pull nextchat:latest

# Run the container
docker run \
  -e OPENAI_API_KEY=sk_xxxxx \
  -e CODE=password123 \
  -p 3000:3000 \
  nextchat:latest
```

--------------------------------

### JSON Configuration Example

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/documentation-structure.md

This is an example of a JSON configuration.

```json
// JSON config
```

--------------------------------

### Mintlify Configuration Example

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/documentation-structure.md

This JSON object shows a complete example of the mint.json configuration file. It includes settings for project name, logos, favicon, brand colors, top bar links and CTA, sidebar anchors, main navigation structure, and footer social links.

```json
{
  "name": "NextChat",                    // Documentation project name
  "logo": {                              // Logo configuration
    "dark": "/logo/dark.svg",           // Used in dark theme
    "light": "/logo/light.svg"          // Used in light theme
  },
  "favicon": "/favicon.svg",            // Browser tab icon
  "colors": {
    "primary": "#315ef8"                // Brand color (blue)
  },
  "topbarLinks": [                       // Top navigation links
    {
      "name": "Support",
      "url": "mailto:contact@nextchat.dev"
    }
  ],
  "topbarCtaButton": {                   // Top-right CTA button
    "name": "Home",
    "url": "https://nextchat.dev"
  },
  "anchors": [                           // Quick access links in sidebar
    {
      "name": "Quick Start",
      "icon": "book-open-cover",
      "url": "/quickstart/quickstart"
    },
    {
      "name": "Community",
      "icon": "discord",
      "url": "https://discord.gg/F7avu4xV4r"
    }
  ],
  "navigation": [                        // Main navigation structure
    {
      "group": "Get Started",
      "pages": [
        "introduction",
        "quickstart/quickstart",
        {
          "group": "Selfhost",
          "pages": [
            "quickstart/selfhost/vercel",
            "quickstart/selfhost/zeabur",
            "quickstart/selfhost/docker",
            "quickstart/selfhost/env"
          ]
        },
        "development"
      ]
    },
    {
      "group": "Model",
      "pages": [
        "models/intro",
        {
          "group": "Providers",
          "pages": [
            "models/openai",
            "models/gemini",
            "models/ollama",
            "models/groq"
          ]
        }
      ]
    }
  ],
  "footerSocials": {                     // Footer social links
    "twitter": "https://twitter.com/NextChatDev",
    "github": "https://github.com/ChatGPTNextWeb"
  }
}
```

--------------------------------

### Bash Command Example

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/documentation-structure.md

This is an example of a bash command.

```bash
# Bash commands
```

--------------------------------

### Start LM Studio Server and Configure NextChat

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/integration-guide.md

Instructions to start an LM Studio server and configure NextChat to use it, including setting API key, base URL, and custom models. Note the specific base URL format for LM Studio.

```bash
# Start LM Studio server (default port 1234)
# Through LM Studio GUI

# NextChat configuration
OPENAI_API_KEY=unused
BASE_URL=http://localhost:1234/v1
CUSTOM_MODELS=+local-model
```

--------------------------------

### Start Local Development Server

Source: https://github.com/chatgptnextweb/docs/blob/main/README.md

Run the Mintlify development server from the root of your documentation directory. Ensure mint.json is present.

```bash
mintlify dev
```

--------------------------------

### Install Mintlify CLI with Yarn

Source: https://github.com/chatgptnextweb/docs/blob/main/development.mdx

Install the Mintlify CLI globally using Yarn. Ensure Node.js version 18.10.0 or higher is installed.

```bash
yarn global add mintlify
```

--------------------------------

### Start vLLM Server and Configure NextChat

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/integration-guide.md

Instructions to start a vLLM server and configure NextChat to use it, including setting API key, base URL, and custom models.

```bash
# Start vLLM server
python -m vllm.entrypoints.openai.api_server --model meta-llama/Llama-2-7b-hf

# NextChat configuration
OPENAI_API_KEY=unused
BASE_URL=http://localhost:8000
CUSTOM_MODELS=+llama2-7b
```

--------------------------------

### Install Mintlify CLI

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Install the Mintlify CLI globally using npm or yarn. This tool is essential for developing and previewing documentation.

```bash
npm install -g mintlify
# OR using yarn
yarn global add mintlify
```

--------------------------------

### Install Mintlify CLI

Source: https://github.com/chatgptnextweb/docs/blob/main/README.md

Install the Mintlify CLI globally using npm. This command is used to manage and preview documentation locally.

```bash
npm i -g mintlify
```

--------------------------------

### Create New MDX Page

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Create a new Markdown Extended (MDX) file for documentation content. This example shows creating a new file for an Anthropic model provider guide.

```bash
# Example: Create a new model provider guide
touch docs/models/anthropic.mdx
Add frontmatter:
```mdx
---
title: "Claude / Anthropic"
description: "Integration with Anthropic's Claude models"
---

# Your content here
```
```

--------------------------------

### Verify Mintlify Installation

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Check if the Mintlify CLI has been installed correctly by running the version command.

```bash
mintlify --version
```

--------------------------------

### Python Code Example

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/documentation-structure.md

This is an example of Python code.

```python
# Python code
```

--------------------------------

### Start LocalAI and Configure NextChat

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/integration-guide.md

Instructions to start LocalAI using Docker and configure NextChat to use it, including setting API key, base URL, and custom models.

```bash
# Start LocalAI
docker run -p 8080:8080 localai/localai

# NextChat configuration
OPENAI_API_KEY=unused
BASE_URL=http://localhost:8080
CUSTOM_MODELS=+gpt4all
```

--------------------------------

### Docker Compose Configuration Example

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/deployment-checklist.md

This is an example of a docker-compose.yml file used for deploying NextChat with Docker. It includes placeholders for environment variables that need to be configured.

```yaml
version: "3.8"

services:
  nextchat:
    image: "nextchat:latest"
    container_name: "nextchat"
    ports:
      - "3000:3000"
    environment:
      # OpenAI API Key (required)
      - OPENAI_API_KEY=
      # Access Password (optional)
      - CODE=
      # Google Gemini API Key (optional)
      - GOOGLE_API_KEY=
      # Azure OpenAI Configuration (optional)
      - AZURE_URL=
      - AZURE_API_KEY=
      - AZURE_API_VERSION=
      # Model Restrictions (optional)
      - DISABLE_GPT4=
      - HIDE_USER_API_KEY=
    restart: unless-stopped
```

--------------------------------

### Vercel Quick Start Deployment Workflow

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/deployment-platforms.md

Recommended workflow for deploying NextChat using Vercel's one-click deployment. Requires authorization and setting environment variables.

```text
1. Click Vercel Deploy Button
   ↓
2. Authorize GitHub
   ↓
3. Add OPENAI_API_KEY
   ↓
4. Add CODE (optional)
   ↓
5. Deploy
   ↓
6. Access via Vercel URL
```

--------------------------------

### Basic OpenAI Setup with Access Control

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/configuration.md

Configure basic OpenAI API key and a single access password. Users must enter the specified password to access the application.

```shell
CODE=password1
OPENAI_API_KEY=sk_xxxxx
```

--------------------------------

### Minimal Environment Configuration

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/README.md

Set the OPENAI_API_KEY environment variable for basic setup. This is the minimum required for many AI providers.

```bash
OPENAI_API_KEY=sk_xxxxx
```

--------------------------------

### List Installed Ollama Models

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/deployment-checklist.md

Command to list all models that have been downloaded and installed via Ollama. This helps in verifying model availability.

```bash
ollama list
```

--------------------------------

### Check Node.js Version

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Verify the installed Node.js version. Mintlify requires version 18.10.0 or higher.

```bash
node --version
```

--------------------------------

### Configure Multiple AI Providers

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/integration-guide.md

Set environment variables to configure multiple AI providers. This example shows how to set up OpenAI, Google Gemini, Azure, and notes for Ollama and custom Groq endpoints.

```bash
# OpenAI as primary
OPENAI_API_KEY=sk_xxxxx
BASE_URL=https://api.openai.com

# Google Gemini as secondary
GOOGLE_API_KEY=google_key_xxxxx

# Azure as tertiary
AZURE_URL=https://resource.openai.azure.com/openai/deployments/gpt4
AZURE_API_KEY=azure_key_xxxxx

# Local Ollama for testing
# Configured via UI settings panel

# Custom Groq endpoint
# Configured via custom endpoint in settings
```

--------------------------------

### Re-install Dependencies

Source: https://github.com/chatgptnextweb/docs/blob/main/README.md

If 'mintlify dev' is not running, use this command to re-install project dependencies. This can resolve issues related to missing or corrupted installations.

```bash
mintlify install
```

--------------------------------

### Install Ollama Model

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/quick-start-reference.md

Troubleshoot the 'No Models Available' error for Ollama by ensuring the desired model is installed using the ollama pull command.

```bash
ollama pull llama2
```

--------------------------------

### Front Matter Example

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/documentation-structure.md

All MDX files begin with YAML front matter to define page metadata like title and description.

```mdx
---
title: "Page Title"
description: "Brief description shown in metadata"
---
```

--------------------------------

### Run Mintlify Development Server on Custom Port

Source: https://github.com/chatgptnextweb/docs/blob/main/development.mdx

Customize the port for the Mintlify development server using the --port flag. This example runs on port 3333.

```bash
mintlify dev --port 3333
```

--------------------------------

### Start NextChat Docker Containers

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/deployment-checklist.md

Command to start the NextChat Docker containers in detached mode using docker-compose. Ensure your docker-compose.yml is correctly configured.

```bash
docker-compose up -d
```

--------------------------------

### Mintlify Port In Use Error

Source: https://github.com/chatgptnextweb/docs/blob/main/development.mdx

Example of an error message when the default port (3000) is already in use.

```text
Error: listen EADDRINUSE: address already in use :::3000
```

--------------------------------

### User Journey 1: SaaS User Navigation Flow

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/documentation-structure.md

Illustrates the navigation path for a SaaS user starting from the introduction to configuring a model provider.

```markdown
introduction
    ↓
quickstart/quickstart (links to SaaS)
    ↓
models/intro (choose provider)
    ↓
models/{provider} (configure provider)
```

--------------------------------

### Local Model Setup with Ollama

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/quick-start-reference.md

Configure NextChat to use local models via Ollama. Set the OpenAI Endpoint to your local Ollama server and specify a custom model.

```bash
# In NextChat settings:
# - OpenAI Endpoint: http://localhost:11434/
# - API Key: (leave empty)
# - Custom Model: llama2
```

--------------------------------

### Integrate Custom Model Provider

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/INDEX.md

Guide to integrating a custom model provider with NextChat. Ensure your endpoint is OpenAI-compatible and configure the necessary environment variables.

```text
1. Read: integration-guide.md → Custom Endpoint Integration
2. Verify custom endpoint is OpenAI-compatible
3. Read: model-providers.md → Custom Endpoints
4. Set BASE_URL to custom endpoint
5. Set CUSTOM_MODELS with provider's model names
6. Test configuration
```

--------------------------------

### Navigate to Documentation Root

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Ensure you are in the root directory of the documentation project, which contains the mint.json configuration file.

```bash
cd /path/to/docs
# Verify mint.json exists
ls -la mint.json
```

--------------------------------

### Development Environment Configuration

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/feature-reference.md

Sets up a development environment enabling shareable URLs, allowing local models alongside GPT-4, and facilitating easy switching between local and cloud models.

```bash
OPENAI_API_KEY=sk_test_xxxxx
DISABLE_FAST_LINK=0
CUSTOM_MODELS=+llama2-local,+mixtral-local,gpt-4=Production GPT-4
```

--------------------------------

### CodeGroup Component for Multiple Languages

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/documentation-structure.md

The CodeGroup component allows displaying equivalent code examples in multiple languages side-by-side.

```mdx
<CodeGroup>
  ```bash npm
  npm install nextchat
  ```
  
  ```bash yarn
  yarn add nextchat
  ```
</CodeGroup>
```

--------------------------------

### Public Deployment Configuration

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/feature-reference.md

Sets up a public deployment requiring a password, limiting models to GPT-3.5 for cost control, and using the server's API key.

```bash
CODE=public_password
OPENAI_API_KEY=sk_xxxxx
DISABLE_GPT4=1
HIDE_USER_API_KEY=1
```

--------------------------------

### Deploy NextChat Quickly

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/INDEX.md

Follow these steps to deploy NextChat using Vercel in under 5 minutes. Ensure you have your OPENAI_API_KEY ready.

```text
1. Read: quick-start-reference.md → Vercel section
2. Click Vercel deploy button
3. Add OPENAI_API_KEY environment variable
4. Click Deploy
5. Test at generated URL
```

--------------------------------

### Pull NextChat Docker Image

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/deployment-checklist.md

Command to pull the latest NextChat Docker image from a registry. This is a prerequisite before starting the containers.

```bash
docker pull nextchat:latest
```

--------------------------------

### Configure Multiple API Providers

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/model-providers.md

Set environment variables to enable multiple AI model providers simultaneously. Users can then select their preferred provider in the UI.

```bash
# Enable both OpenAI and Google
OPENAI_API_KEY=sk_xxxxx
GOOGLE_API_KEY=google_key_123456

# Also add Azure as tertiary
AZURE_URL=https://resource.openai.azure.com/openai/deployments/gpt4
AZURE_API_KEY=azure_key_xxxxx
AZURE_API_VERSION=2022-06-01-preview
```

--------------------------------

### Restart Mintlify Development Server

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Use these commands to reinstall dependencies and restart the development server when Mintlify Dev isn't running.

```bash
mintlify install
mintlify dev
```

--------------------------------

### Mintlify Configuration File

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/MANIFEST.md

The mint.json file is used to configure the Mintlify documentation site, including site structure, navigation, and other build settings.

```json
{
  "name": "nextchat",
  "logo": {
    "light": "./public/logo/light.svg",
    "dark": "./public/logo/dark.svg"
  },
  "favicon": "./public/favicon.svg",
  "colors": {
    "primary": "#0070f3",
    "light": "#e3f2f9",
    "dark": "#0a2540"
  },
  "navigation": [
    {
      "group": "Docs",
      "pages": [
        "introduction",
        "development"
      ]
    },
    {
      "group": "Models",
      "pages": [
        "models/intro",
        "models/openai",
        "models/gemini",
        "models/azure",
        "models/groq",
        "models/ollama"
      ]
    },
    {
      "group": "Quickstart",
      "pages": [
        "quickstart/quickstart",
        "quickstart/selfhost/env",
        "quickstart/selfhost/vercel",
        "quickstart/selfhost/zeabur",
        "quickstart/selfhost/docker"
      ]
    },
    {
      "group": "Contribution",
      "pages": [
        "contribution/overview"
      ]
    }
  ]
}
```

--------------------------------

### Environment Configuration with Access Control

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/README.md

Configure both an access control password (CODE) and an API key for enhanced security. This is useful for protecting your deployment.

```bash
CODE=mypassword
OPENAI_API_KEY=sk_xxxxx
```

--------------------------------

### Basic Docker Run Command

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/deployment-platforms.md

Run a NextChat Docker container with essential environment variables for API keys and access codes. Ensure the image name 'nextchat-image' is correct.

```bash
docker run \
  -e OPENAI_API_KEY=sk_xxxxx \
  -e CODE=password123 \
  -p 3000:3000 \
  nextchat-image
```

--------------------------------

### Team Deployment Configuration with Model Choice

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/feature-reference.md

Configures a team deployment allowing access via multiple passwords, supporting multiple providers, enabling custom models, and allowing users to monitor their balance.

```bash
CODE=team_password1,team_password2
OPENAI_API_KEY=sk_xxxxx
GOOGLE_API_KEY=google_key
CUSTOM_MODELS=gpt-4-turbo=GPT-4 Fast,gpt-3.5-turbo=GPT-3.5 Budget,-gpt-4=GPT-4 Standard
ENABLE_BALANCE_QUERY=1
```

--------------------------------

### Direct API Key Authentication

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/integration-guide.md

Set up direct API key authentication by providing the respective API keys in environment variables for services like OpenAI and Google.

```bash
# Direct API key in environment
OPENAI_API_KEY=sk_xxxxx
GOOGLE_API_KEY=google_key_xxxxx
```

--------------------------------

### Nginx Reverse Proxy Configuration for NextChat

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/deployment-checklist.md

Example Nginx configuration for setting up a reverse proxy to serve NextChat over HTTPS. Ensure you replace placeholders with your actual domain and certificate paths.

```nginx
server {
  listen 443 ssl http2;
  server_name your-domain.com;
  
  ssl_certificate /path/to/cert.pem;
  ssl_certificate_key /path/to/key.pem;
  
  location / {
    proxy_pass http://localhost:3000;
    proxy_set_header Host $host;
    proxy_set_header X-Real-IP $remote_addr;
  }
}
```

--------------------------------

### Configure Ollama CORS and Host via Environment Variables

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/model-providers.md

Set environment variables to configure Ollama's origins and host for serving.

```bash
export OLLAMA_ORIGINS=*://your-domain
export OLLAMA_HOST=0.0.0.0:11434
ollama serve
```

--------------------------------

### NextChat Documentation Directory Structure

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/documentation-structure.md

This snippet displays the hierarchical organization of the NextChat documentation files and directories. It includes configuration files, guides for introduction, development, contribution, and specific model providers, as well as deployment and asset directories.

```tree
docs/
├── mint.json                              # Mintlify configuration
├── README.md                              # Repository readme
├── introduction.mdx                       # Product introduction
├── development.mdx                        # Development setup guide
├── contribution/
│   └── overview.mdx                      # Contribution guidelines
├── models/
│   ├── intro.mdx                         # Model providers intro
│   ├── openai.mdx                        # OpenAI (GPT-3.5, GPT-4) guide
│   ├── gemini.mdx                        # Google Gemini guide
│   ├── azure.mdx                         # Azure OpenAI guide
│   ├── groq.mdx                          # Groq guide
│   └── ollama.mdx                        # Ollama guide
├── quickstart/
│   ├── quickstart.mdx                    # Quick start guide
│   └── selfhost/
│       ├── env.mdx                       # Environment variables reference
│       ├── vercel.mdx                    # Vercel deployment guide
│       └── (docker.mdx planned)          # Docker deployment guide
├── images/                                # Image assets
│   ├── cover.png                         # Cover image
│   ├── selfhost/
│   │   └── vercel-env.png               # Vercel config screenshot
│   └── models/
│       ├── groq/
│       │   ├── config.png               # Groq configuration
│       │   └── groq-api-key.png         # Groq API key setup
│       └── ollama/
│           └── ollama_settings.png      # Ollama settings screenshot
├── logo/
│   ├── dark.svg                          # Dark mode logo
│   └── light.svg                         # Light mode logo
└── favicon.svg                           # Favicon
```

--------------------------------

### Clone Documentation Repository

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Clone the NextChat documentation repository from GitHub and navigate into the cloned directory.

```bash
git clone https://github.com/ChatGPTNextWeb/docs.git
cd docs
```

--------------------------------

### Multiple API Keys and Passwords for Load Balancing

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/configuration.md

Set up multiple OpenAI API keys for load balancing and configure multiple access passwords. Requests are distributed across API keys, and multiple passwords are accepted.

```shell
OPENAI_API_KEY=key1,key2,key3
CODE=secure_pass1,secure_pass2
```

--------------------------------

### Clear Mintlify Cache and Reinstall on Windows

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

If you encounter file path errors on Windows, clear the Mintlify cache, clone the repository, and restart the development server.

```bash
# Delete Mintlify cache
rmdir /s %userprofile%\.mintlify

# Clone Mintlify repository
cd C:\Users\Username\.mintlify\ 
git clone https://github.com/mintlify/mint.git

# Restart
mintlify dev
```

--------------------------------

### Mintlify Configuration Structure

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Defines the overall structure and metadata for your documentation site, including project name, logos, colors, navigation, and social links.

```json
{
  "$schema": "https://mintlify.com/schema.json",
  "name": "NextChat",                    // Project name
  "logo": {
    "dark": "/logo/dark.svg",            // Dark mode logo
    "light": "/logo/light.svg"           // Light mode logo
  },
  "favicon": "/favicon.svg",             // Browser favicon
  "colors": {
    "primary": "#315ef8"                 // Primary brand color
  },
  "topbarLinks": [
    {
      "name": "Support",
      "url": "mailto:contact@nextchat.dev"
    }
  ],
  "topbarCtaButton": {
    "name": "Home",
    "url": "https://nextchat.dev"
  },
  "anchors": [
    {
      "name": "Quick Start",
      "icon": "book-open-cover",
      "url": "/quickstart/quickstart"
    }
  ],
  "navigation": [
    {
      "group": "Get Started",
      "pages": ["introduction", "quickstart/quickstart"]
    }
  ],
  "footerSocials": {
    "twitter": "https://twitter.com/NextChatDev",
    "github": "https://github.com/ChatGPTNextWeb"
  }
}
```

--------------------------------

### Google Gemini and OpenAI Provider Configuration

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/configuration.md

Enable both Google Gemini and OpenAI models by configuring their respective API keys and URLs. Users can select between providers.

```shell
GOOGLE_API_KEY=google_key_123456
GOOGLE_URL=https://googleapis.com/gemini/v1
OPENAI_API_KEY=sk_xxxxx
```

--------------------------------

### Self-Hosted Deployment Workflow

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/deployment-platforms.md

Workflow for self-hosting NextChat using Docker or Kubernetes. Involves cloning the repository, configuring environment variables, and setting up networking and security.

```text
1. Clone Repository (or pull Docker image)
   ↓
2. Configure Environment Variables
   ↓
3. Run Docker Container or Deploy to Kubernetes
   ↓
4. Configure DNS/Custom Domain
   ↓
5. Set up HTTPS (Let's Encrypt or manual cert)
   ↓
6. Monitor and Update
```

--------------------------------

### Enable API Balance Query

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/feature-reference.md

Set the ENABLE_BALANCE_QUERY environment variable to '1' to allow users to view their OpenAI API usage balance in the UI. This feature requires an OpenAI API key with billing enabled.

```bash
ENABLE_BALANCE_QUERY=1    # Enable balance display
ENABLE_BALANCE_QUERY=0    # Disable balance display
# Not set = default (disabled)
```

--------------------------------

### Verify Docker Container Status

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/quick-start-reference.md

Troubleshoot 'Page Won't Load' errors when using Docker by verifying that the NextChat container is running.

```bash
docker ps
```

--------------------------------

### Configure Custom OpenAI-Compatible Endpoint

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/integration-guide.md

Set the API key (if required) and base URL for a custom endpoint, such as vLLM.

```bash
# Custom endpoint (vLLM example)
OPENAI_API_KEY=anything         # Some endpoints don't require a key
BASE_URL=http://localhost:8000
```

--------------------------------

### Configure Proxy Service Endpoint

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/integration-guide.md

Set the API key and base URL for a third-party OpenAI proxy service.

```bash
# Proxy service
OPENAI_API_KEY=sk_xxxxx
BASE_URL=https://api.proxy.com
```

--------------------------------

### No Authentication (Local Ollama)

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/integration-guide.md

Set up local Ollama for use without authentication by setting OPENAI_API_KEY to 'unused' and providing the local base URL.

```bash
# Ollama (no key needed)
OPENAI_API_KEY=unused
BASE_URL=http://localhost:11434
```

--------------------------------

### No Authentication (Local LM Studio)

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/integration-guide.md

Configure LM Studio for local use without authentication by setting OPENAI_API_KEY to 'unused' and specifying the local API endpoint.

```bash
# LM Studio (no key needed)
OPENAI_API_KEY=unused
BASE_URL=http://localhost:1234/v1
```

--------------------------------

### Configure Custom OpenAI-Compatible Endpoints

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/model-providers.md

Set the BASE_URL and OPENAI_API_KEY environment variables to point to a custom OpenAI-compatible API endpoint.

```bash
BASE_URL=http://your-custom-endpoint.com:8000
OPENAI_API_KEY=your-api-key
```

--------------------------------

### Multiple AI Provider Configuration

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/quick-start-reference.md

Set up NextChat to use multiple AI providers by configuring their respective API keys and endpoints.

```bash
OPENAI_API_KEY=sk_xxxxx
GOOGLE_API_KEY=google_key_xxxxx
AZURE_URL=https://resource.openai.azure.com/openai/deployments/gpt4
AZURE_API_KEY=azure_key_xxxxx
```

--------------------------------

### Configure Standard OpenAI Endpoint

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/integration-guide.md

Set the API key and base URL for the default OpenAI endpoint.

```bash
OPENAI_API_KEY=sk_xxxxx
BASE_URL=https://api.openai.com
```

--------------------------------

### Azure OpenAI Deployment URL Format

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/model-providers.md

Illustrates the required format for the Azure OpenAI deployment URL, specifying placeholders for resource and deployment names.

```text
https://{resource-name}.openai.azure.com/openai/deployments/{deployment-name}
```

--------------------------------

### Tip Component Usage

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/documentation-structure.md

The Tip component is ideal for sharing helpful hints and best practices.

```mdx
<Tip>
  Helpful hints and best practices
</Tip>
```

--------------------------------

### Code Blocks in MDX

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Demonstrates how to embed code blocks with syntax highlighting for different languages (bash and Python) within MDX files.

```mdx
```bash
npm install
```

```python
print("Hello, world!")
```
```

--------------------------------

### Basic Markdown Syntax in MDX

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Illustrates basic Markdown elements like headings, bold, italics, and lists within MDX files.

```mdx
# Heading 1
## Heading 2
### Heading 3

**Bold text**
*Italic text*

- List item 1
- List item 2

1. Numbered item 1
2. Numbered item 2
```

--------------------------------

### Update mint.json for Navigation

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Modify the mint.json configuration file to include a newly created page in the documentation's navigation menu.

```json
{
  "navigation": [
    {
      "group": "Model",
      "pages": [
        "models/intro",
        "models/openai",
        "models/gemini",
        "models/anthropic"  // Add new page
      ]
    }
  ]
}
```

--------------------------------

### Push to Fork

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/development-setup.md

Push your local branch to your forked repository.

```bash
git push origin docs/my-feature
```

--------------------------------

### User Journey 2: Self-Hosted User Navigation Flow

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/documentation-structure.md

Details the navigation path for a self-hosted user, from introduction through deployment and AI provider configuration.

```markdown
introduction
    ↓
quickstart/quickstart (choose deployment)
    ↓
quickstart/selfhost/{platform} (deployment guide)
    ↓
quickstart/selfhost/env (configuration)
    ↓
models/{provider} (choose AI provider)
```

--------------------------------

### Update Mintlify CLI to Latest with npm

Source: https://github.com/chatgptnextweb/docs/blob/main/development.mdx

Update the Mintlify CLI to the latest version using npm. This ensures your local website reflects the most recent changes.

```bash
npm i -g mintlify@latest
```

--------------------------------

### Vercel Deployment URL

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/deployment-platforms.md

Use this URL to initiate a one-click deployment of NextChat to Vercel. It pre-configures the repository and common environment variables.

```url
https://vercel.com/new/clone?repository-url=https%3A%2F%2Fgithub.com%2FYidadaa%2FChatGPT-Next-Web&env=OPENAI_API_KEY&env=CODE&env=GOOGLE_API_KEY&project-name=chatgpt-next-web&repository-name=ChatGPT-Next-Web
```

--------------------------------

### Contribute to NextChat Documentation

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/INDEX.md

Instructions for contributing to the NextChat documentation. This involves setting up the development environment, writing, and submitting changes.

```text
1. Read: development-setup.md → Full guide
2. Install Mintlify CLI
3. Clone documentation repository
4. Create new .mdx file
5. Run `mintlify dev`
6. Write and test locally
7. Commit and push
8. Create pull request
```

--------------------------------

### Configure Ollama CORS for Local Access

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/model-providers.md

Run the Ollama server with specific origins allowed for local development.

```bash
OLLAMA_ORIGINS=*://localhost ollama serve
```

--------------------------------

### Check Custom Endpoint Models

Source: https://github.com/chatgptnextweb/docs/blob/main/_autodocs/integration-guide.md

This command queries your custom endpoint to list available models. It's useful for confirming that your endpoint is running and accessible.

```bash
curl http://localhost:8000/v1/models
```