Settings & Configuration

Configure LLM providers, embeddings, and platform connections.

Web LLM Provider

The AI model used for the web chat interface. Choose from 8 providers:

Ollama (local)
OpenAI
Claude
Gemini
Grok
Qwen
Fireworks
Together
i
Ollama runs locally — zero API cost and full privacy. Your code never leaves your computer. Install from ollama.com and run ollama serve. Codeteel auto-discovers available models.

For cloud providers, enter your API key. Keys are encrypted with AES-256-GCM before storage — never stored in plaintext. The UI shows only the first 7 characters.

Platform LLM Provider

A separate LLM configuration for Slack, Telegram, and Discord. Cloud providers only — Ollama cannot work for platforms because the server cannot reach your local machine.

One active platform provider at a time. Required if you want to use platform integrations.

Embedding Provider

Required for code indexing and semantic search. All providers output 1536-dimension vectors.

ProviderModelPrice
OpenAItext-embedding-3-small$0.02/1M tokens
Geminitext-embedding-004$0.015/1M tokens
Mistralmistral-embed$0.10/1M tokens
Voyagevoyage-code-2$0.12/1M tokens
Cohereembed-english-v3.0$0.10/1M tokens