AI Providers

Chatelope connects to various AI providers for chat, image generation, and other features. This guide covers how to set up each provider.

Overview

Chatelope supports two types of AI providers:

1. Cloud Providers - Commercial APIs (OpenAI, Anthropic, Google)
2. Self-Hosted - Run models on your own hardware (Ollama, LM Studio)

Cloud Providers

OpenAI

Available models: GPT-4o, GPT-4 Turbo, GPT-4, GPT-3.5 Turbo

Setup:

1. Get an API key from platform.openai.com
2. Go to Settings > AI Settings
3. Select OpenAI as provider
4. Enter your API key
5. Choose your default model

Additional features:

• Image generation with DALL-E 3
• Text-to-speech with OpenAI TTS
• Transcription with Whisper

Anthropic

Available models: Claude 3.5 Sonnet, Claude 3 Opus, Claude 3 Haiku

Setup:

1. Get an API key from console.anthropic.com
2. Go to Settings > AI Settings
3. Select Anthropic as provider
4. Enter your API key
5. Choose your default model

Google

Available models: Gemini 1.5 Pro, Gemini 1.5 Flash

Setup:

1. Get an API key from aistudio.google.com
2. Go to Settings > AI Settings
3. Select Google as provider
4. Enter your API key
5. Choose your default model

Self-Hosted Providers

For detailed self-hosted setup, see Self-Hosted AI.

Ollama

Run open-source models locally:

1. Install Ollama
2. Pull a model: ollama pull llama3
3. In Chatelope, select Ollama as provider
4. Enter server URL (default: http://localhost:11434)

LM Studio

GUI for running local models:

1. Install LM Studio
2. Download and load a model
3. Start the local server
4. In Chatelope, select LM Studio as provider
5. Enter server URL (default: http://localhost:1234)

Provider Settings

Default Provider

Set your preferred provider for new conversations:

Settings > AI Settings > Default Provider

Per-Conversation Override

Change provider for a specific conversation:

1. Open the conversation
2. Click the gear icon
3. Select a different provider

API Key Security

API keys are stored securely:

• Encrypted at rest
• Never sent to Chatelope servers
• Only transmitted to the provider's API

Model Selection

Automatic Selection

Enable smart model selection based on task:

• Simple queries use faster models
• Complex tasks use more capable models

Manual Selection

Choose specific models for control over:

• Cost - Cheaper models for routine tasks
• Speed - Faster models for quick responses
• Quality - Best models for important work

Cost Tracking

Monitor API usage:

1. Go to Analytics > Usage
2. View costs by provider, model, and time period
3. Set budget alerts

Troubleshooting

Invalid API Key

• Verify the key is correct
• Check the key has proper permissions
• Ensure billing is active on your account

Rate Limits

• Reduce concurrent requests
• Use a model with higher limits
• Upgrade your API plan

Connection Errors

• Check your internet connection
• Verify provider status pages
• Try a different provider