New3 developments · 1 source · within a dayllama.cpp b11998 adds CUDA 13 and Snapdragon supportHow it developedOctober 30, 2026llama.cpp ReleasesWhere it startedllama.cpp b11998 adds CUDA 13 and Snapdragon supportOctober 30, 2026llama.cpp Releasesllama.cpp adds Limite 1B model supportllama.cpp adds native support for the Limite 1B model with core changes across gguf-py and the inference engineOctober 30, 2026llama.cpp Releasesllama.cpp adds OpenCL iq2 quantization for Adrenollama.cpp adds OpenCL iq2 quantization support for Adreno GPUs