workflowstacks
ai-agent

RAFT: Accelerate Machine Learning

Get building blocks for high performance ML apps with RAFT, used by developers.
1,036 stars248 forksCudaGuide quality 8/10Updated 8/22/2026100% free · open source
What it does

RAFT provides building blocks for high performance machine learning applications, including fundamental algorithms and primitives, to help developers write efficient apps more easily

When to use it
  • •When building a high-performance machine learning application that requires GPU acceleration
  • •When needing widely-used algorithms and primitives for machine learning and information retrieval
  • •When developing applications that require CUDA-accelerated computations
Ready-to-paste prompt
To build and run the RAFT example application, use the command: `cd build && cmake .. && make && ./examples/raft_example`
Heads up: RAFT requires a CUDA-compatible NVIDIA GPU to function, so ensure you have a supported GPU installed and the CUDA toolkit is properly configured on your system
Saves to your device
Use with Claude
New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install
Claude Code
mkdir -p ~/.claude/skills/raft && curl -fsSL https://workflowstacks.com/api/skills/raft/claude-skill -o ~/.claude/skills/raft/SKILL.md
Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)
claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How RAFT: Accelerate Machine Learning works
Codeflow
Free to inspect

RAFT: Accelerate Machine Learning is a large Cuda project (~33k lines across 217 code files, plus 42 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, Apache-2.0 license, has a test suite.

Size
Large codebase
~33k lines · 217 code files · ~5 h to skim
Setup
Developer setup
A real software project. Use it via its install path; don't expect to read it all.
Runs on
Python
No API keys detected
Cuda 63%C++ 21%Jupyter Notebook 9%Python 3%Cython 2%
Where to start reading
  1. 1
    README.md
    Start here — what it does and how to install it
  2. 2
    pyproject.toml
    Dependencies and the commands it exposes
  3. 3
    cpp/src/util/memory_pool.cpp
    Inside cpp/src/ — the main logic begins here
What's in each folder
cpp/src/Core code — the actual logic17 files
cpp/internal/Core code — the actual logic2 files
docs/Documentation72 files
cpp/include/Folder422 files
python/Folder93 files
cpp/bench/Folder32 files
ci/Folder18 files
cpp/scripts/Helper scripts10 files
READMEHas testsDocumentedCI checksApache-2.0 licenseUpdated this month
Quick Actions
Details
Creator
NVIDIA
Language
Cuda
Category
ai-agent
Published
6/13/2019

Are you the creator of this tool? Claim your listing → and earn 85% of every sale.

Related skills

More ai-agent tools founders pair with this one.

ai-agent★ 276,806
Track GitHub Stars with 996.ICU
Get a popular GitHub star counter for your startup, backed by 276k+ GitHub stars, ideal for founders.
ai-agent★ 264,191
Superpowers
An agentic skills framework & software development methodology that works.
ai-agent★ 244,424
Linux: Open Source Kernel
Get the Linux kernel source tree for custom development, ideal for startup founders in need of a flexible OS foundation with 236k+ GitHub stars
ai-agent★ 210,750
Vue
This is the repo for Vue 2. For Vue 3, go to https://github.com/vuejs/core
ai-agent★ 204,069
Improve Code with andrej-karpathy-skills
Get better code behavior with andrej-karpathy-skills, 165k+ GitHub stars, for founders using LLMs.
ai-agent★ 197,010
FreeDomain: Free Website Domain
Get a free domain with FreeDomain. For startup founders.