Your full OpenAI-compatible LLM server.
In a single Docker container.
Runs anywhere. Costs nothing.
Chat, completions, time-series analysis, and anomaly scoring — all running locally with zero cloud bills and total data control.
Basilisk is a complete local LLM server + analytics engine, packaged as a single Docker container. Drop it on any machine — laptop, server, edge device — and instantly get OpenAI-compatible chat and completion endpoints, plus deep time-series database analysis and intelligent anomaly scoring.
No cloud. No per-token fees. No data leaving your network. Just raw, private intelligence where you need it.
Drop-in OpenAI-compatible API. Use it with LangChain, LlamaIndex, custom scripts, or any existing OpenAI client — just change the base URL.
Native integration with VictoriaMetrics, Prometheus, and other time-series databases. Ask natural-language questions about your metrics and get real answers.
Intelligent anomaly detection on your time-series data using the LLM + statistical models. Get scored alerts with plain-English explanations.
One command to run. Works on x86_64 and ARM. Runs on laptops, servers, Kubernetes, or even a Raspberry Pi.
Your data never leaves your infrastructure. Perfect for sensitive environments, regulated industries, or anyone tired of cloud bills.
Works with any GGUF model (Llama 3, Mistral, Phi, Gemma, etc.). Swap models by changing one environment variable.
Run powerful local inference with full control over cost, latency, and data sovereignty.
Basilisk is currently in private beta.
If you want early access, a demo, or to run it in your environment — let’s talk.