lucidrains / native-sparse-attention-pytorchLinks
Implementation of the sparse attention pattern proposed by the Deepseek team in their "Native Sparse Attention" paper
β761Updated last month
Alternatives and similar repositories for native-sparse-attention-pytorch
Users that are interested in native-sparse-attention-pytorch are comparing it to the libraries listed below
Sorting:
- π³ Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"β877Updated 6 months ago
- H-Net: Hierarchical Network with Dynamic Chunkingβ739Updated this week
- Muon is an optimizer for hidden layers in neural networksβ1,767Updated 2 months ago
- [ICLR 2025] Official PyTorch Implementation of Gated Delta Networks: Improving Mamba2 with Delta Ruleβ308Updated 2 weeks ago
- [ICLR2025 Spotlightπ₯] Official Implementation of TokenFormer: Rethinking Transformer Scaling with Tokenized Model Parametersβ574Updated 7 months ago
- [NeurIPS 2025 Spotlight] TPA: Tensor ProducT ATTenTion Transformer (T6) (https://arxiv.org/abs/2501.06425)β394Updated last week
- Helpful tools and examples for working with flex-attentionβ991Updated 3 weeks ago
- TransMLA: Multi-Head Latent Attention Is All You Need (NeurIPS 2025 Spotlight)β372Updated last week
- Code for Adam-mini: Use Fewer Learning Rates To Gain More https://arxiv.org/abs/2406.16793β437Updated 4 months ago
- Muon is Scalable for LLM Trainingβ1,318Updated 2 months ago
- Official JAX implementation of Learning to (Learn at Test Time): RNNs with Expressive Hidden Statesβ422Updated last year
- Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Modelsβ832Updated 2 months ago
- β302Updated 5 months ago
- When it comes to optimizers, it's always better to be safe than sorryβ373Updated this week
- Quick implementation of nGPT, learning entirely on the hypersphere, from NvidiaAIβ292Updated 4 months ago
- π₯ A minimal training framework for scaling FLA modelsβ255Updated 3 weeks ago
- β527Updated last week
- Official implementation of "Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding"β482Updated 3 weeks ago
- Implementation of π Ring Attention, from Liu et al. at Berkeley AI, in Pytorchβ540Updated 4 months ago
- Official PyTorch implementation of Learning to (Learn at Test Time): RNNs with Expressive Hidden Statesβ1,254Updated last year
- Efficient triton implementation of Native Sparse Attention.β226Updated 4 months ago
- Lightning Attention-2: A Free Lunch for Handling Unlimited Sequence Lengths in Large Language Modelsβ328Updated 7 months ago
- Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation (NeurIPS 2025)β461Updated last week
- Efficient LLM Inference over Long Sequencesβ391Updated 3 months ago
- Dream 7B, a large diffusion language modelβ984Updated last week
- π Efficient implementations of state-of-the-art linear attention modelsβ3,404Updated last week
- Ring attention implementation with flash attentionβ880Updated 3 weeks ago
- Normalized Transformer (nGPT)β191Updated 10 months ago
- β249Updated 3 months ago
- [ICML2024 (Oral)] Official PyTorch implementation of DoRA: Weight-Decomposed Low-Rank Adaptationβ861Updated last year