CUDA-Practice
CUDA编程练习项目-Hands-on CUDA kernels and performance optimization, covering GEMM, FlashAttention, Tensor Cores, CUTLASS, quantization, KV cache, NCCL, and profiling.
psmarter/CUDA-Practice
View on GitHub ↗CUDA编程练习项目-Hands-on CUDA kernels and performance optimization, covering GEMM, FlashAttention, Tensor Cores, CUTLASS, quantization, KV cache, NCCL, and profiling.
CUDA-Practice ranks 270th of 346 in Training Toolkits, last commit 69 days ago.
Star trajectory
19 days tracked2026-07-02159 stars today2026-07-20
Momentum
- in 30d, below the Training Toolkits median of +0.0%.
Maintenance
Last commit 69 days ago, fresher than the median of 227d.
How it compares in Training Toolkits
CUDA-Practice
159
stars
- 30d
commit 69 days ago
Training Toolkits median
326.5
stars
+0.0% 30d
commit ~227d ago
Leader · TTS
46.1k
stars
+0.2% 30d
commit 799 days ago
Ranked 270th of 346 by stars.
- Language
- Cuda
- License
- MIT License
- Contributors
- 3
- Forks
- 13
- Watchers
- 13
- Open issues
- 0
- Releases
- 0
- Created
- 2026-03-08
- Stars this week
- -2
You've seen where CUDA-Practice stands today.
Get the moves before everyone else: the 5 Training Toolkits repos climbing fastest, every Tuesday.
Free weekly digest. Double opt-in, unsubscribe anytime.