Popular repositories Loading

1. llama-cpp-mtp-turboquant-sm120-blackwell-windows llama-cpp-mtp-turboquant-sm120-blackwell-windows Public

   Windows prebuilt of llama.cpp combining Multi-Token Prediction (MTP) + TurboQuant KV cache compression + native sm_120 (Blackwell consumer GPU, FP4 tensor cores). For RTX 5060 Ti / 5070 / 5080 / 5090.

   8 1

2. beellama.cpp beellama.cpp Public

   Forked from Anbeeld/beellama.cpp

   DFlash & TurboQuant in llama.cpp with up to 3x faster generation and 7.5x more KV cache in same VRAM

   C++ 1

3. knowghost knowghost Public

   AI interview assistant — stealth desktop app with speech recognition, LLM chat, vision, and card storage