# Open benchmarks

Open tests for comparing AI models, with public data or code. Each page explains what the test checks, who can use it, how to run it and how to read the published results.

Models for decisions

## [jevalt-bench](https://eschatialabs.com/benchmarks/jevalt-bench/)

3,116 written situations in English, Turkish and German for testing models that choose among fixed answers and assign a probability to each.

1,600 downloads on Hugging Face, as of October 18, 2026.

- [Open the benchmark](https://eschatialabs.com/benchmarks/jevalt-bench/)

- [Dataset (Hugging Face)](https://huggingface.co/datasets/mertkayacs/jevalt-bench)

- [Code (GitHub)](https://github.com/mertkayacs/jevalt)

- [Project site](https://jevalt.mertkayacs.com)

Models for decisions

## [JevOss](https://eschatialabs.com/benchmarks/jevoss/)

An open test suite that checks how changes to a question affect a model that chooses among fixed answers.

- [Open the benchmark](https://eschatialabs.com/benchmarks/jevoss/)

- [Project site](https://jevoss.mertkayacs.com)

- [Code (GitHub)](https://github.com/mertkayacs/jevoss)

- [Documentation](https://mertkayacs.github.io/jevoss/)

AI task assistants

## [Tholos-Bench](https://eschatialabs.com/benchmarks/tholos-bench/)

160 workspace tasks for testing small AI assistants, scored by whether their completed work matches the request.

170 downloads on Hugging Face, as of October 18, 2026.

- [Open the benchmark](https://eschatialabs.com/benchmarks/tholos-bench/)

- [Dataset (Hugging Face)](https://huggingface.co/datasets/mertkayacs/tholos-bench)

- [Code (GitHub)](https://github.com/mertkayacs/tholos/tree/main/tholos/bench)

- [Project site](https://tholos.mertkayacs.com)

Canonical: https://eschatialabs.com/benchmarks/