HuyNguyen-hust / flash-attn-101Links
☆22Updated last year
Alternatives and similar repositories for flash-attn-101
Users that are interested in flash-attn-101 are comparing it to the libraries listed below
Sorting:
- [ICLR 2025] CAMEx: Curvature-Aware Merging of Experts☆22Updated 6 months ago
- Pioneering in Vietnamese Multimodal Large Language Model☆52Updated 8 months ago
- ☆243Updated 3 months ago
- Implementation of the proposed Adam-atan2 from Google Deepmind in Pytorch☆124Updated 9 months ago
- VIT inference in triton because, why not?☆31Updated last year
- Flash-Muon: An Efficient Implementation of Muon Optimizer☆186Updated 3 months ago
- Memory-Efficient CUDA kernels for training ConvNets with PyTorch.☆42Updated 6 months ago
- ☆72Updated last year
- Implementation of Infini-Transformer in Pytorch☆111Updated 8 months ago
- Some personal experiments around routing tokens to different autoregressive attention, akin to mixture-of-experts☆121Updated 11 months ago
- Implementation of the proposed MaskBit from Bytedance AI☆82Updated 10 months ago
- [ICLR 2025 & COLM 2025] Official PyTorch implementation of the Forgetting Transformer and Adaptive Computation Pruning☆132Updated last week
- Just some miscellaneous utility functions / decorators / modules related to Pytorch and Accelerate to help speed up implementation of new…☆124Updated last year
- Official implementation of the paper: "ZClip: Adaptive Spike Mitigation for LLM Pre-Training".☆133Updated 2 weeks ago
- ☆183Updated 11 months ago
- Implementation of a multimodal diffusion transformer in Pytorch☆104Updated last year
- Load compute kernels from the Hub☆287Updated this week
- WIP☆94Updated last year
- Tiny re-implementation of MDM in style of LLaDA and nano-gpt speedrun☆56Updated 6 months ago
- Code for studying the super weight in LLM☆119Updated 9 months ago
- Landing repository for the paper "Softpick: No Attention Sink, No Massive Activations with Rectified Softmax"☆84Updated last week
- Pytorch implementation of the PEER block from the paper, Mixture of A Million Experts, by Xu Owen He at Deepmind☆128Updated last year
- Implementation of the proposed DeepCrossAttention by Heddes et al at Google research, in Pytorch☆93Updated 6 months ago
- ☆89Updated last year
- Triton implement of bi-directional (non-causal) linear attention☆54Updated 7 months ago
- ☆87Updated 6 months ago
- The official repository for HyperZ⋅Z⋅W Operator Connects Slow-Fast Networks for Full Context Interaction.☆39Updated 5 months ago
- The official implementation of the paper "What Matters in Transformers? Not All Attention is Needed".☆176Updated 5 months ago
- Implementation of TiTok, proposed by Bytedance in "An Image is Worth 32 Tokens for Reconstruction and Generation"☆180Updated last year
- Muon fsdp 2☆44Updated last month