Integrated sovereign AI chatbot and inference platform combining Open WebUI with the local Ollama LLM engine and LiteLLM unified gateway.
All containers in this suite are pre-configured with bridge networking, environment variables, and persistent volume paths.
ollama
Ollama is an open-source, lightweight, and extensible framework for running Large Language Models (LLMs) locally, such as Llama 3, Qwen, Mistral, and DeepSeek.
open-webui
Open WebUI is an extensible, feature-rich, and user-friendly web interface for AI models and LLMs, supporting Ollama and OpenAI-compatible APIs with chat, voice, and RAG capabilities.
litellm
LiteLLM is an AI gateway and LLM proxy that unifies access to 100+ Large Language Models (OpenAI, Gemini, Anthropic, Ollama, Azure, Bedrock, HostYourAI) behind a single OpenAI-compatible API format.
speaches
Speaches is an open-source, OpenAI-compatible Speech-to-Text (faster-whisper) and Text-to-Speech (Kokoro/Piper) server supporting GPU acceleration.
NjordDeploy handles the deployment of this multi-container suite over SSH. Compose templates are rendered and started on Docker or Rootless Podman without host agents.