Turns a HuggingFace link into a ready-to-run local model bundle: compiled C++ kernels (llama.cpp, whisper.cpp) and weights, no Python, so packs stay small (59MB whisper, 602MB qwen3-0.6b). CPU/GPU builds; smoke-tested be…
Turns a HuggingFace link into a ready-to-run local model bundle: compiled C++ kernels (llama.cpp, whisper.cpp) and weights, no Python, so packs stay small (59MB whisper, 602MB qwen3-0.6b). CPU/GPU builds; smoke-tested before release.
Discuss this product