Popular repositories Loading

1. llama.cpp-tq3 llama.cpp-tq3 Public

   Forked from ggml-org/llama.cpp

   llama.cpp fork with TQ3_1S/4S CUDA kernels — 3.5-bit WHT quantization achieving Q4s quality at 10% smaller size. Based on RaBitQ-inspired Walsh-Hadamard transform. Enables 27B models on 16GB GPUs w…

   C++ 158 9

2. Qwen3.6-27b-TQ3_4S Qwen3.6-27b-TQ3_4S Public

   Qwen3.6-26b-TQ3_4S Test

   Python 1 2

3. private-TQ3-4S private-TQ3-4S Public

   C++

4. test_tq3 test_tq3 Public

   JavaScript