Native AI
Built-in intelligence for every Obserra app. Runs fully offline with the rule engine — no API key required. Connect to a local LAN model (Ollama, LM Studio, vLLM), open-source cloud AI (Groq, Together AI), or a private enterprise deployment (Azure OpenAI, on-prem GPT) for enhanced contextual advice.
AI Advisor
Offline (Rule Engine) · Ready
Offline · Rule Engine
Zero dependencies. Works on air-gapped networks. All recommendations are deterministic and auditable. Default mode on every install.
LAN · Local AI
Connect to Ollama (port 11434), LM Studio (port 1234), vLLM, llama.cpp, or any OpenAI-compatible local server. Data never leaves your network.
http://192.168.x.x:11434 · llama3 / mistral / phi3
Open Source Cloud
Groq (ultra-fast inference), Together AI, Hugging Face Inference API, Replicate, or any OpenAI-compatible cloud endpoint with an API key.
Enterprise AI
Azure OpenAI (private deployment), AWS Bedrock via compatibility layer, on-prem GPT-4, or any internal OpenAI-compatible API gateway.