Official PyTorch Implementation of Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention
☆344Oct 11, 2026Updated last week
Alternatives and similar repositories for GatedDeltaNet-2
Users that are interested in GatedDeltaNet-2 are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- [ICLR 2025] Official PyTorch Implementation of Gated Delta Networks: Improving Mamba2 with Delta Rule☆689Mar 13, 2026Updated 7 months ago
- Official repository for Parallax (Parameterized Local Linear Attention)☆70Jul 30, 2026Updated 2 months ago
- Linear Attention Architectures: Mechanisms, Trade-offs, and Cross-Layer Routing (Technical Report)☆39Jul 10, 2026Updated 3 months ago
- high-performance linear attention kernel library built on TileLang☆735Oct 14, 2026Updated last week
- FlashKDA: high-performance Kimi Delta Attention kernels☆1,286Sep 1, 2026Updated last month
- 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.
- Delta Attention Residuals - supplementary code and pretrained models☆43Sep 26, 2026Updated 3 weeks ago
- 🚀 Efficient implementations for emerging model architectures☆5,965Updated this week
- Code for Fast-weight Product Key Memory (FwPKM)☆26Mar 18, 2026Updated 7 months ago
- MoonEP: A Perfectly Balanced Expert Parallelism Library via Dynamic Redundant Experts☆1,189Oct 12, 2026Updated last week
- The Newton-Muon optimizer☆35Jun 5, 2026Updated 4 months ago
- ☆1,619Nov 17, 2025Updated 11 months ago
- Gecko Architecture☆19Jan 13, 2026Updated 9 months ago
- CUDA kernels for linear attention variants, written in CuTe DSL and CUTLASS C++.☆557Updated this week
- ☆289Jun 6, 2025Updated last year
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Quasar Attention Continuous-Time Memory Dynamics for Scaling Models Context to Millions of Tokens☆29Mar 26, 2026Updated 6 months ago
- ☆32Aug 18, 2026Updated 2 months ago
- 赛博苏神：你的 research 好老师☆15Jun 16, 2026Updated 4 months ago
- Mixture-of-experts (MoE) training megakernel for NVL72s☆625Oct 1, 2026Updated 3 weeks ago
- [ACL 2026] Official repo for "Native Hybrid Attention for Efficient Sequence Modeling"☆33Sep 17, 2026Updated last month
- implementations and experimentation on mHC by deepseek - https://arxiv.org/abs/2512.24880☆378Feb 17, 2026Updated 8 months ago
- 🔥 A minimal training framework for scaling FLA models☆423Apr 22, 2026Updated 6 months ago
- ☆431Updated this week
- Official code for HiLS-Attention☆150Aug 13, 2026Updated 2 months ago
- Serverless GPU API endpoints on Runpod - Get Bonus Credits • AdSkip the infrastructure headaches. Auto-scaling, pay-as-you-go, no-ops approach lets you focus on innovating your application.
- Official Codebase: LT2: Linear-Time Looped Transformers.☆65Jul 27, 2026Updated 2 months ago
- Attention variant with per-channel multiplicative decay☆50Jun 3, 2026Updated 4 months ago
- DeltaProduct is a new linear recurrent neural network architecture that uses products of generalized Householder matrices as state-transi…☆18Oct 13, 2025Updated last year
- Official repository for paper Routing-Free Mixture-of-Experts.☆19Aug 23, 2026Updated 2 months ago
- ☆90Aug 19, 2026Updated 2 months ago
- Code for Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers (https://arxiv.org/abs/2606.31779)☆53Jul 20, 2026Updated 3 months ago
- Fused MoE kernel☆23Aug 10, 2026Updated 2 months ago
- ☆23Jul 10, 2026Updated 3 months ago
- Official JAX implementation of End-to-End Test-Time Training for Long Context☆713Feb 15, 2026Updated 8 months ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- LM engine is a library for pretraining/finetuning LLMs☆197Sep 25, 2026Updated 3 weeks ago
- Code for ICML 2026 Oral paper "OPUS: Towards Efficient and Principled Data Selection in Large Language Model Pre-training in Every Iterat…☆36May 30, 2026Updated 4 months ago
- Code and paper reproductions for MLPs Are Hebbians☆34Jul 22, 2026Updated 3 months ago
- ☆102Oct 12, 2026Updated last week
- [ICLR 2025 & COLM 2025] Official PyTorch implementation of the Forgetting Transformer and Adaptive Computation Pruning☆154Feb 25, 2026Updated 7 months ago
- Cuda kernels for leveraging LLM sparsity to improve throughput and decrease the memory requirements during inference and training.☆263Jun 29, 2026Updated 3 months ago
- An hardware-aware Efficient Implementation for "Mixture-of-Depths Attention".☆280May 6, 2026Updated 5 months ago