gau-nernst / quantized-training
Explore training for quantized models
☆17Updated 2 months ago
Alternatives and similar repositories for quantized-training:
Users that are interested in quantized-training are comparing it to the libraries listed below
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆74Updated this week
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆103Updated 8 months ago
- extensible collectives library in triton☆84Updated this week
- Boosting 4-bit inference kernels with 2:4 Sparsity☆72Updated 7 months ago
- ☆76Updated 4 months ago
- Fast Hadamard transform in CUDA, with a PyTorch interface☆165Updated 10 months ago
- ☆68Updated 2 months ago
- An experimental CPU backend for Triton (https//github.com/openai/triton)☆40Updated 2 weeks ago
- PyTorch bindings for CUTLASS grouped GEMM.☆77Updated 5 months ago
- ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization☆105Updated 5 months ago
- Fast low-bit matmul kernels in Triton☆275Updated this week
- Fast Matrix Multiplications for Lookup Table-Quantized LLMs☆236Updated last month
- Framework to reduce autotune overhead to zero for well known deployments.☆63Updated this week
- FlexAttention w/ FlashAttention3 Support☆26Updated 5 months ago
- ☆63Updated last week
- ☆103Updated 7 months ago
- Standalone Flash Attention v2 kernel without libtorch dependency☆108Updated 6 months ago
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.☆107Updated this week
- Collection of kernels written in Triton language☆117Updated last month
- A bunch of kernels that might make stuff slower 😉☆29Updated this week
- ☆26Updated 2 weeks ago
- ☆60Updated 3 months ago
- ☆27Updated 2 months ago
- llama INT4 cuda inference with AWQ☆53Updated 2 months ago
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆91Updated this week
- ☆113Updated last week
- ☆157Updated last year
- Ahead of Time (AOT) Triton Math Library☆56Updated 2 weeks ago
- CUDA and Triton implementations of Flash Attention with SoftmaxN.☆68Updated 10 months ago
- PyTorch extension for emulating FP8 data formats on standard FP32 Xeon/GPU hardware.☆107Updated 4 months ago