What is a #cuda MCP server?
An MCP server tagged #cuda implements the Model Context Protocol so AI assistants like Claude, Cursor, and VS Code can access cuda-related tools, data, or APIs.
3 #cuda MCP Servers & Clients
Every MCP server and client below is tagged #cuda — install one to give Claude, Cursor, VS Code, or any other MCP-compatible client access to cuda tools.
All
Ingero
ingero-io
eBPF-based GPU causal observability agent with MCP server. Traces CUDA Runtime/Driver APIs and host kernel events to build causal chains explaining GPU latency
Added 6 months ago
Forge - GPU Kernel Optimization
RightNow-AI
Turn slow PyTorch into fast CUDA/Triton kernels. 32 parallel swarm agents optimize your code on real datacenter GPUs (B200, H200, H100, A100) with up to 14x speedup over torch.compile.
Added 7 months ago
CKG-NVIDIA- AI - Nvidia Developer Stack As A Compressed Knowledge Graph - 20 Domains, 998 Nodes, Agents Traverse Type Dependency Graphs
Yarmoluk
NVIDIA AI developer stack as a Compressed Knowledge Graph (CKG) — 20 domains, 998 nodes, every prerequisite chain declared as typed edges. Agents traverse REQUIRES/ENABLES relationships instead of scanning docs. 4× F1 vs RAG, 11× fewer tokens. No API key required
Added 2 months ago
Frequently asked questions
Common questions about MCP servers and clients tagged #cuda
How many #cuda MCP servers and clients are there?
mcp.so currently lists 3 MCP servers and clients tagged #cuda.
How do I install a #cuda MCP server?
Open any server below and copy its install snippet into Claude Desktop, Cursor, VS Code, or another MCP client's configuration — remote servers need no separate download.