zhuzilin / flash-attention-with-sinkLinks
☆39Updated 2 months ago
Alternatives and similar repositories for flash-attention-with-sink
Users that are interested in flash-attention-with-sink are comparing it to the libraries listed below
Sorting:
- DeeperGEMM: crazy optimized version☆71Updated 5 months ago
- ☆64Updated 5 months ago
- ☆50Updated 4 months ago
- Debug print operator for cudagraph debugging☆13Updated last year
- An experimental communicating attention kernel based on DeepEP.☆34Updated 2 months ago
- Quantized Attention on GPU☆44Updated 10 months ago
- DLSlime: Flexible & Efficient Heterogeneous Transfer Toolkit☆66Updated last week
- ☆115Updated last month
- ☆30Updated 3 months ago
- Accelerate LLM preference tuning via prefix sharing with a single line of code☆43Updated 3 months ago
- ☆53Updated last week
- Estimate MFU for DeepSeekV3☆24Updated 9 months ago
- ☆33Updated last week
- A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆60Updated last month
- ☆99Updated 4 months ago
- Framework to reduce autotune overhead to zero for well known deployments.☆84Updated 2 weeks ago
- Tile-based language built for AI computation across all scales☆66Updated last week
- Code for paper: [ICLR2025 Oral] FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference☆142Updated 4 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆123Updated 4 months ago
- 16-fold memory access reduction with nearly no loss☆105Updated 6 months ago
- A Suite for Parallel Inference of Diffusion Transformers (DiTs) on multi-GPU Clusters☆50Updated last year
- Implement Flash Attention using Cute.☆96Updated 9 months ago
- a simple API to use CUPTI☆11Updated last month
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆132Updated 2 weeks ago
- Utility scripts for PyTorch (e.g. Memory profiler that understands more low-level allocations such as NCCL)☆55Updated 3 weeks ago
- ☆82Updated 8 months ago
- [COLM 2024] SKVQ: Sliding-window Key and Value Cache Quantization for Large Language Models☆24Updated last year
- ☆98Updated last year
- ☆78Updated 5 months ago
- [DAC'25] Official implement of "HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference"☆72Updated 3 months ago