Janus embeds llama.cpp directly inside a lightweight Go binary with a Vulkan acceleration bridge, eliminating the Python runtime and Docker virtualization layers used by typical local LLM stacks. The single-file gateway combines hot-swappable GGUF model execution with native reasoning tag extraction and targets AMD, Intel, and NVIDIA hardware.
Janus strips away Python runtime bloat and Docker virtualization, embedding llama.cpp directly inside a lightweight Go binary with a Vulkan acceleration bridge. By uniting hot-swappable GGUF model execution with native reasoning tag extraction, it offers a single-file gateway for local inference across AMD, Intel, and NVIDIA hardware.
EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.