2 projects
inferhost
Run any Hugging Face GGUF model on your own GPU — chat/vision, text-to-speech (OuteTTS), and image generation (SDXL/Flux/Z-Image/Qwen-Image via stable-diffusion.cpp), all behind one OpenAI-compatible endpoint. Pulls official upstream binaries, no compiling. Type `inferhost` and you're done.
cadence-core
Neural model for next clinical event prediction from EHR sequences using the Narrative Velocity framework
