Triton kernels and PyTorch ops for Block Attention Residuals (AttnRes)
☆89May 29, 2026Updated 3 months ago
Alternatives and similar repositories for flash-attention-residuals
Users that are interested in flash-attention-residuals are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Official repository Flash Local Linear Attention☆40May 28, 2026Updated 3 months ago
- Muon in Int8 Precision Made Possible☆20Jun 18, 2026Updated 3 months ago
- FA4-based Relative Attention Kernel developed by TML and Colfax☆18Sep 11, 2026Updated last week
- Official repository for Parallax (Parameterized Local Linear Attention)☆69Jul 30, 2026Updated last month
- DeeperGEMM: crazy optimized version☆85May 5, 2025Updated last year
- GPU virtual machines on DigitalOcean Gradient AI • AdGet to production fast with high-performance AMD and NVIDIA GPUs you can spin up in seconds. The definition of operational simplicity.
- ☆98Sep 14, 2026Updated last week
- FlashKDA: high-performance Kimi Delta Attention kernels☆1,259Sep 1, 2026Updated 3 weeks ago
- ☆22May 5, 2025Updated last year
- ☆37Dec 31, 2025Updated 8 months ago
- The Newton-Muon optimizer☆33Jun 5, 2026Updated 3 months ago
- Delta Attention Residuals - supplementary code and pretrained models☆43May 20, 2026Updated 4 months ago
- ☆32Jul 2, 2025Updated last year
- Benchmark tests supporting the TiledCUDA library.☆19Nov 19, 2024Updated last year
- ☆54May 19, 2025Updated last year
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- CUDA kernels for linear attention variants, written in CuTe DSL and CUTLASS C++.☆551Updated this week
- An experimental communicating attention kernel based on DeepEP.☆34Jul 29, 2025Updated last year
- ☆37Aug 7, 2025Updated last year
- [MLSys 26] 🥇 Solution for Gated Delta Net Track of MLSys 26 Flash infer competition☆36May 22, 2026Updated 4 months ago
- Fast Polar Decomposition for Muon☆184Jul 2, 2026Updated 2 months ago
- high-performance linear attention kernel library built on TileLang☆706Updated this week
- CODA: Rewriting Transformer Blocks as GEMM-Epilogue Programs☆254Updated this week
- Debug print operator for cudagraph debugging☆18Aug 2, 2024Updated 2 years ago
- Parallel Associative Scan for Language Models☆18Jan 8, 2024Updated 2 years ago
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- ☆87Aug 19, 2026Updated last month
- Triton kernels for dynamic causal short convolutions.☆30Jun 4, 2026Updated 3 months ago
- Open implementation of Attention Residuals (Kimi Team, arXiv:2603.15031)☆83Apr 30, 2026Updated 4 months ago
- A dynamic binary instrumentation tool for tracing and analyzing GPU kernel instructions.☆79Updated this week
- Handwritten GEMM using Intel AMX (Advanced Matrix Extension)☆17Jan 11, 2025Updated last year
- ☆255Nov 19, 2025Updated 10 months ago
- FLA but cuTile☆27Apr 17, 2026Updated 5 months ago
- Accelerating MoE with IO and Tile-aware Optimizations☆769Aug 29, 2026Updated 3 weeks ago
- Skills for writing tilelang and debugging with CUDA toolkits.☆144Sep 15, 2026Updated last week
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- ☆51Aug 17, 2026Updated last month
- ☆73Mar 24, 2026Updated 6 months ago
- ☆39Dec 14, 2025Updated 9 months ago
- AdaSplash: Adaptive Sparse Flash Attention (aka Flash Entmax Attention)☆49May 20, 2026Updated 4 months ago
- Spectral Sphere Optimizer☆133Mar 23, 2026Updated 6 months ago
- An efficient implementation of the NSA (Native Sparse Attention) kernel☆135Jun 24, 2025Updated last year
- NCU-driven iterative optimization workflow for CUDA/CUTLASS/Triton/CuTe DSL kernels.☆25Apr 10, 2026Updated 5 months ago