Tools overview (ML pipeline and our tools)

Modified on Fri, 13 Oct 2023 at 06:38 AM

Tools overview in aiXplain: Building and enhancing AI systems

aiXplain is a platform that helps users to effortlessly construct, diagnose, and enhance AI systems. In this article, we will introduce you to the main tools that aiXplain provides and how you can leverage them to achieve your AI goals.

Discover: models and datasets

Discover serves as aiXplain's marketplace, showcasing a variety of models and datasets from multiple suppliers. It enables users to explore, evaluate, and select the best-fitting assets based on specific criteria such as price, popularity, and use-case suitability.

Key features of Discover include:

- Functions: Navigate through a wide range of models and datasets that cater to various domains and applications. Filter by model types, ASR, TTS etc. 

- Trial and comparison: Experiment with different models, assess their outputs, and make informed decisions.

- Design: Bring models into Design and start building your own AI applications. 

- Custom asset request: If you can't find what you need, request a custom asset, and aiXplain will develop it for you.

Discover is your marketplace to a rich repository of AI assets, enabling easy access and utilization in your projects.

Benchmark: Assess model performance

Benchmark is aiXplain's tool for evaluating the performance of AI models comprehensively. This tool offers in-depth analysis and actionable insights into various aspects of model quality, latency, footprint, cost, and bias.

Key functionalities of Benchmark include:

- Continuous comparison: Measure and compare your AI system's performance against other models or your expectations over time.

-  Custom testing: Test your uploaded models against existing or custom datasets.

-  Detailed reports: Access detailed reports and visualizations that illuminate the strengths and weaknesses of your models.

Benchmark empowers both suppliers to refine their models and users to identify the most suitable model for their unique requirements.

FineTune: Enhance pre-trained models

FineTune is aiXplain's no-code tool that enables you to refine pre-trained models using new data without starting from scratch. This tool streamlines the process of improving model accuracy and performance.

Key aspects of FineTune include:

-  Data-driven optimization: FineTune pre-trained models with your data to enhance accuracy.

-  Minimal training time: Achieve more accurate models with reduced training time.

-  Customization: Tailor pre-trained models to your data for optimized results.

FineTune facilitates effortless model enhancement, allowing you to create more accurate models in less time.

Design: Build AI pipelines

Design is aiXplain's intuitive canvas tool for constructing intricate AI pipelines. It helps you to build customized pipelines with ease, integrating various models, datasets, and components to achieve specific objectives.

Prominent features of Design are:

-  Drag-and-drop interface: Easily assemble components with a simple drag-and-drop mechanism.

-  Pipeline customization: Create pipelines that suit your needs by integrating models, datasets, preprocessing, and more.

-  No-code logic: Construct pipelines without requiring extensive coding knowledge.

-  API integration: Deploy pipelines with API endpoints seamlessly.

Design is the tool for constructing tailored AI pipelines that cater to your unique requirements.


aiXplain is a platform that offers a range of tools that help you create and maintain AI systems easily. Whether you are looking for ready-to-use AI assets, performance evaluation, AutoML, fine-tuning, or pipeline building, aiXplain has the right tool for you.

If you have any further questions, please don't hesitate to reach out to our customer support team.

Contact us at

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select atleast one of the reasons

Feedback sent

We appreciate your effort and will try to fix the article