RWKV/rwkv.cpp

INT4/INT5/INT8 and FP16 inference on CPU for RWKV language model

45
/ 100
Emerging

This project helps developers run large language models (LLMs) like RWKV on standard computer CPUs or even specialized GPUs with better speed and reduced memory usage. It takes an existing RWKV model file, converts it into an optimized format, and then allows for text generation or chatbot interactions directly on your machine. This is designed for software developers or machine learning engineers who need to deploy and run LLMs efficiently on diverse hardware, especially those with limited GPU resources.

1,563 stars. No commits in the last 6 months.

Use this if you are a developer looking to deploy and run RWKV-based large language models efficiently on CPU-centric systems or need to optimize model size and inference speed.

Not ideal if you are a non-technical end-user looking for a ready-to-use application or service, as this requires development setup and coding knowledge.

large-language-models on-device-AI model-optimization AI-inference edge-computing
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 19 / 25

How are scores calculated?

Stars

1,563

Forks

125

Language

C++

License

MIT

Last pushed

Mar 23, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/RWKV/rwkv.cpp"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.