neural-compression
llama2-chatbot-cpu
A LLaMA2-7b chatbot with memory running on CPU, and optimized using smooth quantization, 4-bit quantization or Intel® Extension For PyTorch with bfloat16.
- 15
- Python
- MIT
- 3 y ago
LVQAC
[CVPR 2023] Official PyTorch implementation of paper "LVQAC: Lattice Vector Quantization Coupled With Spatially Adaptive Companding for Efficient Learned Image Compression".
- 7
- Python
- BSD-3-Clause
- 2 wk ago
StateSMix
Lossless compressor in pure C - a Mamba state-space model trained online, sparse n-gram context mixing (2- to 32-gram) and arithmetic coding. No GPU, no pre-trained weights, no dependencies — beats xz -9e by up to 8.7% on files to 10 MB. arXiv:2605.02904
- 6
- C
- Apache-2.0
- 3 mo ago
OLVQ
[NeurIPS 2024] Official PyTorch implementation of paper "Learning Optimal Lattice Vector Quantizers for End-to-end Neural Image Compression".
- 7
- Python
- BSD-3-Clause
- 2 wk ago
bitgolf
Neural text compressor in Rust: a byte-level int8 transformer and range coder, bit-exact on x86-64 and WebAssembly, 40% smaller than brotli on web prose.
- 0
- Rust
- MIT
- 3 wk ago


