Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Training
☆222Aug 19, 2024Updated last year
Alternatives and similar repositories for LightSeq
Users that are interested in LightSeq are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Python package for rematerialization-aware gradient checkpointing☆27Oct 31, 2023Updated 2 years ago
- Ring attention implementation with flash attention☆998Sep 10, 2025Updated 6 months ago
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆653Jan 15, 2026Updated 2 months ago
- Large Context Attention☆769Oct 13, 2025Updated 5 months ago
- ☆46Nov 10, 2023Updated 2 years ago
- Virtual machines for every use case on DigitalOcean • AdGet dependable uptime with 99.99% SLA, simple security tools, and predictable monthly pricing with DigitalOcean's virtual machines, called Droplets.
- Zero Bubble Pipeline Parallelism☆452May 7, 2025Updated 10 months ago
- Reference implementation of "Softmax Attention with Constant Cost per Token" (Heinsen, 2024)☆24Jun 6, 2024Updated last year
- Memory optimization and training recipes to extrapolate language models' context length to 1 million tokens, with minimal hardware.☆755Sep 27, 2024Updated last year
- Code for the paper "Rethinking Benchmark and Contamination for Language Models with Rephrased Samples"☆320Dec 20, 2023Updated 2 years ago
- Distributed Compiler based on Triton for Parallel Systems☆1,394Mar 11, 2026Updated 2 weeks ago
- Linear Attention Sequence Parallelism (LASP)☆88Jun 4, 2024Updated last year
- FlexAttention w/ FlashAttention3 Support☆27Oct 5, 2024Updated last year
- PyTorch bindings for CUTLASS grouped GEMM.☆185Feb 19, 2026Updated last month
- Official repository for LongChat and LongEval☆533May 24, 2024Updated last year
- DigitalOcean Gradient AI Platform • AdBuild production-ready AI agents using customizable tools or access multiple LLMs through a single endpoint. Create custom knowledge bases or connect external data.
- [ICML'24] Data and code for our paper "Training-Free Long-Context Scaling of Large Language Models"☆449Oct 16, 2024Updated last year
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆211Sep 21, 2024Updated last year
- Triton-based implementation of Sparse Mixture of Experts.☆270Oct 3, 2025Updated 5 months ago
- ☆84Dec 2, 2022Updated 3 years ago
- Serving multiple LoRA finetuned LLM as one☆1,148May 8, 2024Updated last year
- Implementation of Hyena Hierarchy in JAX☆10Apr 30, 2023Updated 2 years ago
- Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training☆1,864Updated this week
- Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads☆2,722Jun 25, 2024Updated last year
- Standalone Flash Attention v2 kernel without libtorch dependency☆112Sep 10, 2024Updated last year
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- FlashInfer: Kernel Library for LLM Serving☆5,194Updated this week
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on H…☆3,231Updated this week
- YaRN: Efficient Context Window Extension of Large Language Models☆1,686Apr 17, 2024Updated last year
- Microsoft Automatic Mixed Precision Library☆635Dec 1, 2025Updated 3 months ago
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,273Aug 28, 2025Updated 6 months ago
- ☆78May 4, 2021Updated 4 years ago
- [NeurIPS'23] H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models.☆507Aug 1, 2024Updated last year
- Distributed IO-aware Attention algorithm☆24Sep 24, 2025Updated 6 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆147May 29, 2025Updated 9 months ago
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,324Mar 6, 2025Updated last year
- Training and serving large-scale neural networks with auto parallelization.☆3,187Dec 9, 2023Updated 2 years ago
- Best practice for training LLaMA models in Megatron-LM☆663Jan 2, 2024Updated 2 years ago
- Byted PyTorch Distributed for Hyperscale Training of LLMs and RLs☆1,000Mar 3, 2026Updated 3 weeks ago
- An Attention Superoptimizer☆22Jan 20, 2025Updated last year
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆1,041Sep 4, 2024Updated last year
- Tile primitives for speedy kernels☆3,244Mar 17, 2026Updated last week