These tools and metrics are designed to help AI actors develop and use trustworthy AI systems and applications that respect human rights and are fair, transparent, explainable, robust, secure and safe.
TrustyAI Explainability Toolkit
TrustyAI Explainability is an open-source Java toolkit for Explainable AI (XAI), released under the Apache-2.0 license. It provides fairness metrics and explainability algorithms that can be applied to a model's inputs and outputs without requiring access to the model's internal structure. The toolkit is distributed as a Java library and as a standalone containerised REST service, with companion Python bindings and a Kubernetes Operator for deployment alongside models running in a cluster.
The core library implements the fairness metrics and explainability algorithms, which can be run directly against a model's predictions. A separate connectors module queries live models served through Kubernetes-native frameworks such as KServe or ModelMesh over gRPC, and an Arrow-based module transfers data between the Java core and a separate TrustyAI Python library. The toolkit has been tested against decision-model formats (DMN, PMML) and NLP models through integration with the Kogito engine. These components can be used individually as a library call or combined into an operationalised service that generates fairness scores and explanations for models running in production.
Auto-discovered on 2026-08-31 by OECD Catalogue Automation
About the tool
You can click on the links to see the associated tools
Tool type(s):
Objective(s):
Purpose(s):
Country/Territory of origin:
Lifecycle stage(s):
Type of approach:
Maturity:
Usage rights:
License:
Target users:
Required skills:
Risk management stage(s):
Technology platforms:
Github stars:
- 63
Github forks:
- 51
Use Cases
Would you like to submit a use case for this tool?
If you have used this tool, we would love to know more about your experience.
Add use case
Partnership on AI
