zhuzilin / ring-flash-attention
Ring attention implementation with flash attention
β721Updated last month
Alternatives and similar repositories for ring-flash-attention:
Users that are interested in ring-flash-attention are comparing it to the libraries listed below
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inferenceβ463Updated 2 weeks ago
- Large Context Attentionβ696Updated 2 months ago
- Implementation of π Ring Attention, from Liu et al. at Berkeley AI, in Pytorchβ509Updated 5 months ago
- Zero Bubble Pipeline Parallelismβ379Updated 3 weeks ago
- Microsoft Automatic Mixed Precision Libraryβ587Updated 6 months ago
- Official repository for LightSeq: Sequence Level Parallelism for Distributed Training of Long Context Transformersβ207Updated 7 months ago
- π Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flashβ¦β234Updated this week
- Memory optimization and training recipes to extrapolate language models' context length to 1 million tokens, with minimal hardware.β706Updated 6 months ago
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3.β1,113Updated last week
- Fast inference from large lauguage models via speculative decodingβ700Updated 7 months ago
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Headsβ442Updated last month
- [NeurIPS'23] H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models.β433Updated 8 months ago
- π³ Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"β601Updated 2 weeks ago
- PyTorch bindings for CUTLASS grouped GEMM.β111Updated 3 months ago
- β311Updated last year
- A collection of memory efficient attention operators implemented in the Triton language.β257Updated 9 months ago
- A PyTorch Native LLM Training Frameworkβ763Updated 3 months ago
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decodingβ1,221Updated 3 weeks ago
- A throughput-oriented high-performance serving framework for LLMsβ785Updated 6 months ago
- Efficient LLM Inference over Long Sequencesβ365Updated last month
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inferenceβ265Updated 4 months ago
- LLM KV cache compression made easyβ444Updated 2 weeks ago
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Seβ¦β624Updated 3 weeks ago
- Super-Efficient RLHF Training of LLMs with Parameter Reallocationβ255Updated 2 months ago
- REST: Retrieval-Based Speculative Decoding, NAACL 2024β198Updated 4 months ago
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.β783Updated 6 months ago
- [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Trainingβ170Updated this week
- [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantizationβ337Updated 7 months ago
- flash attention tutorial written in python, triton, cuda, cutlassβ315Updated 2 months ago
- FlagScale is a large model toolkit based on open-sourced projects.β257Updated this week