Back to List
Notice:This resource is provided by a third-party author. Please review the code with AI tools or manually before use to ensure security and compatibility.
Cudamirage-project/mirage

mirage

Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

73.4/100
2.4KForks: 233
View on GitHubHomepage →
Loading report...

Similar Projects

how-to-optim-algorithm-in-cuda

60

how to optimize some algorithm in cuda.

Cuda3.1K

rtp-llm

80

RTP-LLM: Alibaba's high-performance LLM inference engine for diverse applications.

Cuda1.3K

raft

79

RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.

Cuda1.0K

SpargeAttn

50

[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.

Cuda1.0K
Back to List