ByteDance-Seed / cudaLLMLinks
☆125Updated 4 months ago
Alternatives and similar repositories for cudaLLM
Users that are interested in cudaLLM are comparing it to the libraries listed below
Sorting:
- ☆116Updated 7 months ago
- ☆65Updated 8 months ago
- Quantized Attention on GPU☆44Updated last year
- Implement Flash Attention using Cute.☆100Updated last year
- ☆84Updated 11 months ago
- ☆52Updated 7 months ago
- DeeperGEMM: crazy optimized version☆74Updated 8 months ago
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆242Updated last month
- NVIDIA cuTile learn☆144Updated 3 weeks ago
- PyTorch bindings for CUTLASS grouped GEMM.☆135Updated 7 months ago
- Autonomous GPU Kernel Generation via Deep Agents☆197Updated last week
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆141Updated 7 months ago
- Distributed MoE in a Single Kernel [NeurIPS '25]☆171Updated this week
- Code for paper: [ICLR2025 Oral] FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference☆160Updated 2 months ago
- ☆96Updated 9 months ago
- Fast and memory-efficient exact kmeans☆131Updated last month
- A Suite for Parallel Inference of Diffusion Transformers (DiTs) on multi-GPU Clusters☆53Updated last year
- [HPCA 2026] A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆73Updated 2 weeks ago
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆151Updated 3 months ago
- ☆35Updated 9 months ago
- Framework to reduce autotune overhead to zero for well known deployments.☆91Updated 3 months ago
- Odysseus: Playground of LLM Sequence Parallelism☆79Updated last year
- Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.☆46Updated 6 months ago
- [DAC'25] Official implement of "HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference"☆95Updated 3 weeks ago
- ☆133Updated 7 months ago
- ☆39Updated 4 months ago
- 16-fold memory access reduction with nearly no loss☆109Updated 9 months ago
- QuTLASS: CUTLASS-Powered Quantized BLAS for Deep Learning☆155Updated last month
- Triton adapter for Ascend. Mirror of https://gitee.com/ascend/triton-ascend☆97Updated this week
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆123Updated last year