Alternatives to NVIDIA/cutlass
10.7k stars · C++ · back to the repository
Two measured signals: repositories that share topics (at least two), and repositories discussed in the same threads. Cards are ranked by stars gained this week.
Shares topics with
Repositories with at least two topics in common: most shared topics first. Topics are labels maintainers put on their own repositories, so they say what a project claims to be, not what its code does.
NVIDIA/nccl
Optimized primitives for collective multi-GPU communication
Shared: cpp, cuda, deep-learning, gpu +1 more
CVCUDA/CV-CUDA
CV-CUDA™ is an open-source, GPU accelerated library for cloud-scale image processing and computer vision.
Shared: cpp, cuda, gpu, nvidia +1 more
parca-dev/parca-agent
eBPF based always-on CPU/GPU profiler auto-discovering targets in Kubernetes and systemd, zero code changes or restarts needed!
Shared: cpp, cuda, gpu, nvidia +1 more
NVIDIA/nvshmem
NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process communication and coordination overheads by allowing programmers to perform one-sided communication from within CUDA kernels and on CUDA streams.
Shared: cpp, cuda, deep-learning, nvidia +1 more
NVIDIA/warp
A Python framework for GPU-accelerated simulation, robotics, and machine learning.
Shared: cuda, gpu, nvidia, python
NVIDIA/TransformerEngine
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance with lower memory utilization in both training and inference.
Shared: cuda, deep-learning, gpu, python
NVIDIA/cudnn-frontend
cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.
Shared: cuda, deep-learning, gpu, nvidia
Competes with
Repositories in the same collection, of a comparable size (within 5× in stars) and from a different owner, that were discussed in the same threads. Same-owner projects are excluded because they are usually pieces of the same tool, not rivals. A repository also stays out if the two never met in a discussion.
Mentioned together in 1 thread · on the same trending list 1 round.
Compare side by side
Compare head to head: cutlass vs wasmtime
Everything we measure about NVIDIA/cutlass · stars over time, growth, contributors, issues and pull requests.