Offline-first desktop app and server for local LLMs, TTS, STT, and image generation with voice chat and workflow builder.
AI Runner is an all-in-one, offline-first desktop application, headless server, and Python library for local LLMs, TTS, STT, and image generation. It supports real-time voice conversations, custom AI agents, drag-and-drop LangGraph workflow builder, and image generation with Stable Diffusion and FLUX models. Runs locally with no external APIs by default, ensuring privacy.
check_circleOllama and OpenAI compatible endpoints
check_circleMulti-language support (English, Japanese, Spanish, French, Chinese, Korean)
check_circleDocker deployment
Use Cases
lightbulbDevelopers run a local LLM server with OpenAI-compatible API to integrate AI chat into their applications without cloud costs.
lightbulbContent creators generate images from text prompts using Stable Diffusion models, iterating quickly with inpainting and LoRA support.
lightbulbResearchers build custom AI agents with configurable personalities and RAG-enhanced knowledge for domain-specific Q&A.
lightbulbTeams automate workflows by designing drag-and-drop LangGraph pipelines that execute LLM calls, image generation, and TTS in sequence.
lightbulbPrivacy-conscious users run all AI models offline, ensuring no data leaves their machine while enjoying voice conversations and chatbot interactions.
lightbulbLanguage learners practice conversations with an LLM that supports multiple languages, receiving real-time TTS responses in their target language.
lightbulbSystem administrators deploy a headless server on a remote machine to provide AI services (LLM, TTS, STT) to other applications via REST API.
