ONNX Runtime
onnxruntime.ai
Cross-platform engine for running ML models.
Quick Information
- Category:
- MLOps & Infrastructure
- Pricing:
- Free
- Platforms:
- Windows, macOS, Linux, Android, iOS
- Free Plan:
- Yes
- API Available:
- Yes
- Last Verified:
- October 2026
What Is ONNX Runtime?
ONNX Runtime is a cross-platform engine for running machine learning models, maintained by Microsoft. It is free and the licence is MIT, according to a Libraries.io page for the PyPI package, though that page covers version 1.26.0 and I did not confirm the licence on the newest release. The latest release I found is v1.31.0, shown as latest on the microsoft/onnxruntime GitHub releases page, with Maven listing 1.31.0 on October 10, 2026. Its notes say it stabilises the model-package and EPContext data APIs and improves CPU model loading and quantized mixture-of-experts inference. Some pages are stale: Snyk shows 1.30.0 as latest and piwheels lists 1.27.0 as its newest entry.
Don't confuse the runtime with the ONNX model format, which is a separate project whose licence Wikipedia says started as MIT and later moved to Apache 2.0. ONNX Runtime has no price, and its value is running one exported model across CPUs, GPUs and accelerators through execution providers, so you can avoid maintaining separate stacks. Vendors such as Intel and Qualcomm ship their own forks or integrations, which showed up in my search results, but they are not the official project. Because it is a runtime, the cost is the hardware you deploy on.
ONNX Runtime suits developers shipping models to servers, desktops, mobile and the edge. Compare OpenVINO, TensorRT-LLM and Triton Inference Server. Browse the MLOps & Infrastructure tools directory for more.
What Can You Use ONNX Runtime For?
Cross-platform inference
One model, many targets.
Edge and mobile deployment
Execution providers.
ONNX Runtime Features
- v1.31.0 (October 2026) — Latest found.
- Model-package and EPContext APIs — Stabilised.
- Quantized MoE inference — Improved.
Who Is ONNX Runtime For?
- Application developers — Portable inference.
- Edge and mobile teams — Lightweight runtime.
ONNX Runtime Pros & Limitations
What We Like
- Free and MIT-licensed
- Broad hardware support
- Microsoft-backed
Limitations
- Needs ONNX export
- Performance depends on the execution provider
- Licence on latest release not checked
ONNX Runtime Pricing
| Plan | Price | Main Features |
|---|---|---|
| Open source | Free (MIT) | Licence confirmed only on an older version |
Pricing was last verified October 2026. Check the official website for the latest pricing.
AI Listing Jungle Ranking
- Global Rank:
- #1138
- Overall Score:
- 26.3/100
- Monthly Users:
- –
- Rank Change:
- ▲ 433
This ranking reflects AI Listing Jungle's own independent scoring methodology and is not an endorsement by ONNX Runtime or its company. How Our Rankings Work
Frequently Asked Questions
Is it free?
Yes.
Latest version?
v1.31.0 in October 2026.
Licence?
MIT per Libraries.io.
Same as ONNX?
No; ONNX is the model format.
Our Take
The portable inference workhorse.
Pick the right execution provider.
ONNX Runtime Alternatives
ONNX Runtime Comparisons
Ready to try ONNX Runtime?
Visit the official website to get started.
Visit Official WebsiteAI Listing Jungle is an independent AI tools directory. Product names, logos, and trademarks belong to their respective owners.