Oxide is a lightweight containerized model inferencing server built without unnecessary frameworks! 🦀
Oxide Home Page
- Inferences custom models on any hardware you want.
- Choose between custom tokenizers, full control over your model.
- incredibly lightweight
Installing Oxide can be done from source.
Clone this repository,
git clone https://github.com/dawi369/Oxide.git
To render run server, navigate to the root of the repo directory, and run
cargo run --release
Oxide: Inference your custom models on the cloud!
Usage:
curl -X POST http://<your_FQDS_or_localhost>>/inference \
-H "Content-Type: text/plain" \
-d "Please generate a warm welcome."
(c) David Erwin