Practical overview
When it fits
Use the repository as a historical model catalog and documentation resource; follow its notice to Hugging Face when seeking models originally hosted there.
Who it is for
- Developers exploring pretrained ONNX models for inference experiments
- Researchers reviewing model examples and their supplied test data
Documented uses
- Browse models across vision, language, speech, and other documented task areas.
- Use model test inputs and reference outputs to compare results from an ONNX-capable backend.
- Consult model usage documentation and inspect architectures with the suggested visualization tool.
Requirements & limits
The repository is public and licensed under Apache-2.0. Its notice says model files would no longer be available for LFS download starting July 1, 2025, and directs readers to Hugging Face for models originally hosted in the repository.
- Model packages can include sample inputs and reference outputs in protobuf or NumPy formats.
- Running the examples requires an ONNX-capable inference backend and adaptation of the README’s starter interface.
- The repository is preserved for historical purposes, and its notice redirects model-file access to Hugging Face.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Validate an ONNX model with its sample data
Use a model package’s test inputs and reference outputs to compare results from an ONNX-capable inference backend.
Before you start
- A model package containing model.onnx and test data in protobuf or NumPy format
- Python, NumPy, ONNX utilities, and an inference backend that supports ONNX
Steps
Obtain an available model and its accompanying test data; the repository notice directs users to the ONNX Model Zoo on Hugging Face for models originally hosted there.
Load the model and sample inputs using the format-specific approach shown in the repository README.
Run the model with an ONNX-capable backend, then compare its outputs with the supplied reference outputs.
Expected outcome
A comparison of backend outputs with the model package’s reference outputs for the supplied test data.
Guide limitations
- The README presents starter code that must be adapted to the chosen backend’s model-evaluation interface.
- The repository notice says LFS downloads from this repository would stop starting July 1, 2025.
Compare options for Model inference and serving
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| ONNX Model Zoo | Browse models across vision, language, speech, and other documented task areas. | The repository is public and licensed under Apache-2.0. Its notice says model files would no longer be available for LFS download starting July 1, 2025, and directs readers to Hugging Face for models originally hosted in the repository. | |
| Pulse Of Motion | Estimate a video's physical frame rate from its visual dynamics. | Linux | The repository is public. Its README describes a released checkpoint that downloads from Hugging Face on first inference, with a manual download option. The strongest internal model is not fully released; the README directs interested users to contact the project about access. No license information is provided in the repository evidence. |
| Neural Image Assessment | Train the model using AVA images and rating annotations. | Cloud, Linux, macOS, Windows |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 14 October 2026
- Original website title
- onnx/models · Official homepage title · retained in its source language