andrewkchan / yalm
Yet Another Language Model: LLM inference in C++/CUDA, no libraries except for I/O
☆282Updated 3 months ago
Alternatives and similar repositories for yalm:
Users that are interested in yalm are comparing it to the libraries listed below
- Perplexity GPU Kernels☆235Updated 2 weeks ago
- Flash Attention in ~100 lines of CUDA (forward pass only)☆779Updated 3 months ago
- A throughput-oriented high-performance serving framework for LLMs☆794Updated 7 months ago
- Cataloging released Triton kernels.☆217Updated 3 months ago
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆639Updated last month
- Dynamic Memory Management for Serving LLMs without PagedAttention☆348Updated this week
- Fast low-bit matmul kernels in Triton☆288Updated this week
- Fastest kernels written from scratch☆223Updated 2 weeks ago
- Materials for learning SGLang☆379Updated last month
- ☆199Updated last week
- A low-latency & high-throughput serving engine for LLMs☆341Updated this week
- flash attention tutorial written in python, triton, cuda, cutlass☆328Updated 3 months ago
- A curated collection of resources, tutorials, and best practices for learning and mastering NVIDIA CUTLASS☆161Updated last month
- Applied AI experiments and examples for PyTorch☆261Updated last month
- CUDA/Metal accelerated language model inference☆540Updated last week
- Efficient LLM Inference over Long Sequences☆368Updated this week
- KV cache compression for high-throughput LLM inference☆126Updated 2 months ago
- BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.☆587Updated 2 months ago
- Puzzles for learning Triton, play it with minimal environment configuration!☆290Updated 4 months ago
- ☆207Updated 2 months ago
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆248Updated 5 months ago
- CPU inference for the DeepSeek family of large language models in C++☆288Updated this week
- Repository for the QUIK project, enabling the use of 4bit kernels for generative inference - EMNLP 2024☆179Updated last year
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆803Updated 7 months ago
- a minimal cache manager for PagedAttention, on top of llama3.☆82Updated 7 months ago
- Distributed Triton for Parallel Systems☆415Updated 2 weeks ago
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆304Updated 9 months ago
- A Easy-to-understand TensorOp Matmul Tutorial☆342Updated 7 months ago
- [ICLR'25] Fast Inference of MoE Models with CPU-GPU Orchestration☆206Updated 5 months ago
- Mirage: Automatically Generating Fast GPU Kernels without Programming in Triton/CUDA☆805Updated this week