Load and execute ONNX models in Rust with the ort crate.
npx skills add https://github.com/HopeArtOrg/hope-re --skill onnx-rust-integration
Please help me install this Agent Skill. Skill: onnx-rust-integration Source: https://github.com/HopeArtOrg/hope-re/tree/main/.gemini/skills/onnx-rust-integration Command: npx skills add https://github.com/HopeArtOrg/hope-re --skill onnx-rust-integration
This Skill helps developers load and run ONNX models within Rust applications using the ort crate, enabling the use of pre-trained machine learning models in Rust projects.
Load the ONNX model 'noise_algorithm.onnx' and run an inference on the input data 'image.png'.
High-intent search queries and answers about installing and using this skill.
To load and run ONNX models in Rust, use the ort crate to initialize a session, resolve the model path, and execute inference on input data using platform-specific execution providers.
Yes, the ort crate supports hardware acceleration for ONNX inference through platform-specific execution providers including CUDA, DirectML, CoreML, XNNPACK, and CPU.
The best way to resolve ONNX model paths in Rust is using a prioritized resolution strategy that checks downloaded models, bundled resources, and dev-mode paths sequentially.
Yes, you can pass extra index inputs when running ONNX inference in Rust, allowing you to supplement the primary input data with additional tensors required by the model.
Yes, you need platform-specific execution libraries to use ONNX in Rust, as the ort crate relies on execution providers like CUDA, DirectML, CoreML, XNNPACK, or CPU to run models.
acornproverInstall ONNX Runtime static libraries for the ort crate in sandboxed environments.
NVIDIA-OmniverseReads USD prim attributes into CPU or GPU tensors via read_attribute and read_array_attribute.
ai-enhanced-engineerManage TensorFlow.js tensor lifecycle and normalize Teachable Machine inputs in React Native.
JNZader-VaultDeploys ML models across Python, Rust, and Go using ONNX Runtime with quantization and graph transformations for edge scenarios.
