Multi-LLM Providers - Teleton Agent Documentation

Supported Providers

Provider	Models	Best For
`anthropic`	Claude 3.5, Claude 3	Complex reasoning, tool use
`openai`	GPT-4o, GPT-4	General tasks
`groq`	Llama 3, Mixtral	Fast inference, cost-effective
`ollama`	Any local model	Privacy, offline use

Anthropic (Default)

config.yaml

llm:
  provider: anthropic
  model: claude-sonnet-4-20250514
  apiKey: ${ANTHROPIC_API_KEY}
  maxTokens: 4096

OpenAI

config.yaml

llm:
  provider: openai
  model: gpt-4o
  apiKey: ${OPENAI_API_KEY}

Groq

Ultra-fast inference with open-source models:

config.yaml

llm:
  provider: groq
  model: llama-3.1-70b-versatile
  apiKey: ${GROQ_API_KEY}

Ollama (Local)

Run models locally for privacy and offline use:

config.yaml

llm:
  provider: ollama
  model: llama3.1
  baseUrl: http://localhost:11434

Hardware requirements

Local models require sufficient hardware. Recommended: 16GB+ RAM for 7B models, 32GB+ for larger models.

Fallback Configuration

Configure fallback providers for reliability:

config.yaml

llm:
  provider: anthropic
  model: claude-sonnet-4-20250514
  apiKey: ${ANTHROPIC_API_KEY}
  fallback:
    provider: groq
    model: llama-3.1-70b-versatile
    apiKey: ${GROQ_API_KEY}