BabyBench 2025 Competition

BabyBench is a multimodal benchmark of infant behaviors for developmental artificial intelligence. The objective is to teach typical developmental behaviors to MIMo, a multimodal infant model. We provide the embodiment, the simulation environments, and the evaluation metrics; all you need is to implement your ideas.

The first BabyBench Competition will take place at the IEEE ICDL 2025 Conference in Prague! The topic of this first edition will be how infants discover their own bodies. Can you help MIMo learn two typical infant behaviors: self touch and hand regard?

Download the code Learn more

Important dates

• Submission deadline: August 15, 2025
• Finalists announcement: September 1, 2025
• Winner announcement at ICDL: September 19, 2025

---

---

BibTex

@misc{lopez2025babybench,
      title={BabyBench: A Multimodal Benchmark of Infant Behaviors for Developmental AI},
      author={Francisco M. López and Valentin Marcel and Xavier Hinaut and Jochen Triesch and Matej Hoffmann},
      year={2025},
}

---

 

 

---

Next page: About