RTen

Rust-based neural network runtime for running ONNX models efficiently on CPU.

RTen is a lightweight, end-to-end Rust inference runtime that can load ONNX models directly and execute them efficiently on CPU targets. It emphasizes portability, small footprint, and practical deployment across native and WebAssembly environments.

  Version Dockerfile Date Score Supported Ops
RTen
07/27/2026 07:19:10 UTC
0.00%
-

Backend unit tests list

 
RTen

Operators coverage

 
RTen