Efficient LLM Inference over Long Sequences
☆393Jun 25, 2025Updated 8 months ago
Alternatives and similar repositories for Star-Attention
Users that are interested in Star-Attention are comparing it to the libraries listed below
Sorting:
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads☆527Feb 10, 2025Updated last year
- [NeurIPS'24 Spotlight, ICLR'25, ICML'25] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention…☆1,190Sep 30, 2025Updated 5 months ago
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆255Feb 13, 2026Updated 3 weeks ago
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆817Mar 6, 2025Updated last year
- A throughput-oriented high-performance serving framework for LLMs☆947Oct 29, 2025Updated 4 months ago
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆143Dec 4, 2024Updated last year
- Code for paper: [ICLR2025 Oral] FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference☆161Oct 13, 2025Updated 4 months ago
- Implement Flash Attention using Cute.☆102Dec 17, 2024Updated last year
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆374Jul 10, 2025Updated 7 months ago
- Implementation from scratch in C of the Multi-head latent attention used in the Deepseek-v3 technical paper.☆18Jan 15, 2025Updated last year
- [COLM 2024] TriForce: Lossless Acceleration of Long Sequence Generation with Hierarchical Speculative Decoding☆277Aug 31, 2024Updated last year
- A low-latency & high-throughput serving engine for LLMs☆482Jan 8, 2026Updated last month
- 🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"☆969Feb 5, 2026Updated last month
- Benchmark tests supporting the TiledCUDA library.☆18Nov 19, 2024Updated last year
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆21Feb 9, 2026Updated 3 weeks ago
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆1,025Sep 4, 2024Updated last year
- LLM KV cache compression made easy☆936Feb 23, 2026Updated last week
- We invite you to visit and follow our new repository at https://github.com/microsoft/TileFusion. TiledCUDA is a highly efficient kernel …☆194Jan 28, 2025Updated last year
- Ring attention implementation with flash attention☆987Sep 10, 2025Updated 5 months ago
- ☆118May 19, 2025Updated 9 months ago
- [ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-t…☆3,192Jan 17, 2026Updated last month
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆150May 10, 2025Updated 9 months ago
- BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.☆753Aug 6, 2025Updated 7 months ago
- Quantized Attention on GPU☆44Nov 22, 2024Updated last year
- ☆13Jan 7, 2025Updated last year
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆336Jul 2, 2024Updated last year
- FlexAttention w/ FlashAttention3 Support☆27Oct 5, 2024Updated last year
- FlashInfer: Kernel Library for LLM Serving☆5,057Updated this week
- ☆302Jul 10, 2025Updated 7 months ago
- Framework to reduce autotune overhead to zero for well known deployments.☆97Sep 19, 2025Updated 5 months ago
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆106Jun 28, 2025Updated 8 months ago
- 🚀 Efficient implementations of state-of-the-art linear attention models☆4,474Updated this week
- VPTQ, A Flexible and Extreme low-bit quantization algorithm☆675Apr 25, 2025Updated 10 months ago
- [ICML 2024] KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache☆358Nov 20, 2025Updated 3 months ago
- ☆85Apr 18, 2025Updated 10 months ago
- [ICML 2025] XAttention: Block Sparse Attention with Antidiagonal Scoring☆269Jul 6, 2025Updated 8 months ago
- [ICML'24] Data and code for our paper "Training-Free Long-Context Scaling of Large Language Models"☆448Oct 16, 2024Updated last year
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆644Jan 15, 2026Updated last month
- ☆131Nov 11, 2024Updated last year