ELS-RD/transformer-deploy
An inference server that optimizes and deploys Hugging Face Transformer models for production with up to 10X speedup.
Not currently ranked — collecting fresh signals.
star history
Transformer-deploy provides an efficient, scalable inference server for running Hugging Face Transformer models on CPU and GPU. It leverages ONNX Runtime and Nvidia Triton inference server to accelerate model inference compared to standard PyTorch + FastAPI stacks. The tool offers single-command deployment and supports enterprise-grade production scenarios with semantic search and re-ranking use cases.
Frequently asked
- What is ELS-RD/transformer-deploy?
- An inference server that optimizes and deploys Hugging Face Transformer models for production with up to 10X speedup.
- Is transformer-deploy open source?
- Yes — ELS-RD/transformer-deploy is open source, released under the Apache-2.0 license.
- What language is transformer-deploy written in?
- ELS-RD/transformer-deploy is primarily written in Python.
- How popular is transformer-deploy?
- ELS-RD/transformer-deploy has 1.7k stars on GitHub.
- Where can I find transformer-deploy?
- ELS-RD/transformer-deploy is on GitHub at https://github.com/ELS-RD/transformer-deploy.