andrewkchan / yalmLinks
Yet Another Language Model: LLM inference in C++/CUDA, no libraries except for I/O
☆388Updated last month
Alternatives and similar repositories for yalm
Users that are interested in yalm are comparing it to the libraries listed below
Sorting:
- Perplexity GPU Kernels☆395Updated last month
- Materials for learning SGLang☆481Updated last week
- A curated collection of resources, tutorials, and best practices for learning and mastering NVIDIA CUTLASS☆195Updated 2 months ago
- kernels, of the mega variety☆441Updated last month
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆717Updated 4 months ago
- Fast low-bit matmul kernels in Triton☆330Updated this week
- Efficient LLM Inference over Long Sequences☆382Updated 2 weeks ago
- A throughput-oriented high-performance serving framework for LLMs☆840Updated this week
- Cataloging released Triton kernels.☆242Updated 6 months ago
- ☆225Updated this week
- LLM KV cache compression made easy☆535Updated this week
- CUDA/Metal accelerated language model inference☆594Updated last month
- Fastest kernels written from scratch☆290Updated 3 months ago
- Applied AI experiments and examples for PyTorch☆281Updated last month
- Flash Attention in ~100 lines of CUDA (forward pass only)☆867Updated 6 months ago
- Dynamic Memory Management for Serving LLMs without PagedAttention☆401Updated last month
- KernelBench: Can LLMs Write GPU Kernels? - Benchmark with Torch -> CUDA problems☆468Updated this week
- BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.☆643Updated last week
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆855Updated 10 months ago
- KV cache store for distributed LLM inference☆288Updated last month
- a minimal cache manager for PagedAttention, on top of llama3.☆93Updated 10 months ago
- ☆195Updated 2 months ago
- ☆110Updated 3 months ago
- A low-latency & high-throughput serving engine for LLMs☆388Updated last month
- ☆173Updated this week
- A safetensors extension to efficiently store sparse quantized tensors on disk☆135Updated this week
- A Quirky Assortment of CuTe Kernels☆281Updated this week
- An experimental CPU backend for Triton☆135Updated last month
- CPU inference for the DeepSeek family of large language models in C++☆306Updated last month
- Mirage: Automatically Generating Fast GPU Kernels without Programming in Triton/CUDA☆1,540Updated last week