This repositories contains the reference implementation for the Sparse Delta Memory paper.More precisely, it contains the model definition as well as triton and cuda kernels for the Sparse Delta Memory layer.
☆24Jul 9, 2026Updated this week
Alternatives and similar repositories for sparse-delta-memory
Users that are interested in sparse-delta-memory are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- ☆77May 29, 2026Updated last month
- Multi-modal, multi-task modeling of the mouse visual cortex☆16May 5, 2026Updated 2 months ago
- ☆19Dec 12, 2023Updated 2 years ago
- Official Project Page for HLA: Higher-order Linear Attention (https://arxiv.org/abs/2510.27258)☆102Jun 15, 2026Updated 3 weeks ago
- Expanding linear RNN state-transition matrix eigenvalues to include negatives improves state-tracking tasks and language modeling without…☆22Mar 15, 2025Updated last year
- Bare Metal GPUs on DigitalOcean Gradient AI • AdPurpose-built for serious AI teams training foundational models, running large-scale inference, and pushing the boundaries of what's possible.
- ☆20Updated this week
- The Newton-Muon optimizer☆30Jun 5, 2026Updated last month
- Attention variant with per-channel multiplicative decay☆48Jun 3, 2026Updated last month
- Official repository for Parallax (Parameterized Local Linear Attention)☆63Jul 7, 2026Updated last week
- Flash-Linear-Attention models beyond language☆21Aug 28, 2025Updated 10 months ago
- 更纯粹、更高压缩率的Tokenizer in Rust☆14Dec 21, 2024Updated last year
- Training-free KV cache compression via E8 lattice VQ. 2-bit KV that preserves retrieval (30/30 NIAH vs TurboQuant 0/30). Calibration-free…☆25Updated this week
- Aurora optimizer release☆150May 27, 2026Updated last month
- [KernelGYM & Dr. Kernel] A distributed GPU environment and a collection of RL training methods to support RL for Kernel Generations [ICML…☆191Mar 29, 2026Updated 3 months ago
- 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.
- Implementation and explorations into DiscoRL, Discovering state-of-the-art reinforcement learning algorithms, David Silver's last work at…☆20Jun 13, 2026Updated last month
- A Multi-Policy, Multi-Agent RL Training Framework☆30Jun 16, 2026Updated 3 weeks ago
- Flash Attention in 300-500 lines of CUDA/C++☆39Aug 22, 2025Updated 10 months ago
- XWikisCorpus, cross-lingual summarisation, multi-lingual summarisation, pre-trained language models, zero-shot and few-shot summarisation…☆10Nov 4, 2022Updated 3 years ago
- ☆25Dec 29, 2018Updated 7 years ago
- STABILIZING GRADIENTS FOR DEEP NEURAL NETWORKS VIA EFFICIENT SVD PARAMETERIZATION☆16Jun 5, 2018Updated 8 years ago
- Experiments on the impact of depth in transformers and SSMs.☆43Oct 23, 2025Updated 8 months ago
- This is the official implementation for paper "On Powerful Ways to Generate: Autoregression, Diffusion, and Beyond".☆23Nov 17, 2025Updated 7 months ago
- mHC-lite: You Don’t Need 20 Sinkhorn-Knopp Iterations☆91Jan 12, 2026Updated 6 months ago
- 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.
- ☆13Sep 27, 2022Updated 3 years ago
- This is the official code implementation of Bongard-OpenWorld (ICLR 2024).☆14Jan 6, 2025Updated last year
- ☆45Nov 1, 2025Updated 8 months ago
- Efficient PScan implementation in PyTorch☆17Jan 2, 2024Updated 2 years ago
- AGaLiTe: Approximate Gated Linear Transformers for Online Reinforcement Learning (Published in TMLR)☆24Oct 15, 2024Updated last year
- Transformers components but in Triton☆34May 9, 2025Updated last year
- Fast Punctuation Restoration using Transformer Models for Vietnamese☆11Jun 10, 2022Updated 4 years ago
- ☆20Dec 24, 2024Updated last year
- An efficient implementation of the NSA (Native Sparse Attention) kernel☆133Jun 24, 2025Updated last year
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Julia implementation of flash-attention operation for neural networks.☆11May 31, 2023Updated 3 years ago
- ☆12Sep 26, 2019Updated 6 years ago
- Compositional Muon release☆23Jun 5, 2026Updated last month
- 🔥 A minimal training framework for scaling FLA models☆401Apr 22, 2026Updated 2 months ago
- Official code for the paper "Attention as a Hypernetwork"☆58Feb 24, 2026Updated 4 months ago
- Surgical GPU kernel benchmark: 7 hard problems, frontier coding agents, roofline-graded against hardware peak.☆19Jun 12, 2026Updated last month
- Supporting code for the EMNLP 2019 paper "Answers Unite! Unsupervised Metrics for Reinforced Summarization Models"☆14Jun 12, 2023Updated 3 years ago