igel
Train, test, and use machine learning models without writing a single line of code.
Train, test, and optimize models using simple YAML or JSON configuration files.
Igel is a delightful machine learning tool designed to let you build models without writing code. Data scientists and developers who want to bypass boilerplate code or quickly prototype models will find it exceptionally useful.
What it is
It wraps around existing ML frameworks to handle regression, classification, and clustering tasks through simple configuration files. It also offers auto-ML features that can process raw folders of images or text and find the best performing model automatically.
Who it's for
It suits data scientists, developers, and non-technical users needing fast prototypes or proofs of concept. It does not suit advanced researchers who need low-level custom architecture design within training loops.
Key features
Zero Code Training
Train and evaluate models using simple YAML or JSON configuration files.
AutoML Capabilities
Automatically process raw data folders and find the best model for your specific task.
Extensive Model Support
Supports a vast array of regression, classification, and clustering algorithms out of the box.
Flexible Data Inputs
Accepts CSV, TXT, Excel, JSON, HTML files, or raw directory structures.
Up and running in 3 steps
Our honest verdict
What's great
- Builds machine learning prototypes rapidly without writing boilerplate code.
- Supports hyperparameter search and cross-validation natively.
- Offers an optional GUI desktop application (Igel-UI) for ease of use.
- Handles multiple dataset formats including raw image and text folders.
Where it falls short
- Auto-ml features can be computationally expensive as they test multiple models.
- Requires you to structure configuration files accurately in YAML or JSON.
- Not suitable for building entirely novel neural network architectures from scratch.
Pricing & license
Free under its open source license. The only cost is running it yourself on your own hardware.
Open remote AI-training roles
880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
These reviewed AI tools do the same job instantly — from our scored directory.
Use it, or have us set it up
FAQ
Do I need to know Python to use Igel?
No, Igel is designed for zero-code usage where you configure tasks via YAML or JSON files.
Can I use Igel for image classification?
Yes, Igel supports automated image and text classification tasks directly from raw folder structures.
