LLMs & Model Routing - Negentro Docs

Provider	Supported Models	Recommended For
OpenAI	gpt-4o , gpt-4o-mini , o3-mini	Fast, accurate atomic fact extraction.
Anthropic	claude-3-7-sonnet , claude-3-5-haiku	High-fidelity reasoning and complex entity resolution.
Google Gemini	gemini-2.0-flash , gemini-1.5-pro	High-speed inference and multimodal context analysis.
DeepSeek	deepseek-v3 , deepseek-r1	Cost-effective open-weights inference.
Ollama (Local)	llama3.3 , mistral-small , qwen2.5	Air-gapped, zero-cost private deployments.

Supported Providers

Local Ollama Configuration

Run 100% private, on-premise memory pipelines without sending tokens to third-party APIs:

python_ollama.py

from negentro import Memory

config = {
    "llm": {
        "provider": "ollama",
        "config": {
            "model": "llama3.3:latest",
            "base_url": "http://localhost:11434"
        }
    }
}

m = Memory.from_config(config)