Pinned Loading

1. cactus cactus Public

   Quantization, kernels, runtime and inference engine for mobiles, wearables, smart home and robots.

   C++ 5.5k 451

2. needle needle Public

   Foundation model for tiny devices; 14mb, 26m params, 1-6k toks/sec on mobiles, wearables smart home and robots.

   Python 3.3k 254

3. cactus-hybrid cactus-hybrid Public

   On-device models that know when they're wrong: every answer carries a confidence score for cloud handoff.

   Shell 194 3