PeopleofSuperintelligence

Researcher

Jean-Baptiste Alayrac

Jean-Baptiste Alayrac works on models that connect images, video and text, from early video-learning research to the Gemini family. He graduated from Ecole polytechnique and Telecom ParisTech in 2015, and his Inria research focused on structured learning from video and natural language. Flamingo, a visual language model for few-shot learning, appeared in 2022 with him as first author. He was co-lead for multimodal vision on the Gemini report, which was submitted in December 2023. As of October 2026, his listed affiliation is Google in Mountain View.

Now at
Google
Works
65
Citations
6,144
h-index
26

Publication numbers from OpenAlex, as of 15 October 2026.

Timeline

  1. 2015

    A master's in computer vision

    He graduates from Ecole polytechnique and Telecom ParisTech in 2015 and completes a master's in mathematics, machine learning and computer vision.

  2. 2017

    Learning from video and text

    He co-writes Learning from Video and Text via Large-Scale Discriminative Clustering, a paper submitted to arXiv in July 2017 that is to appear at ICCV 2017. At Inria, his work focuses on structured learning from video and natural language.

  3. 2022

    Flamingo for few-shot learning

    He is first author of Flamingo: a Visual Language Model for Few-Shot Learning, submitted to arXiv in April 2022. DeepMind's blog introduces Flamingo in April 2022 as a single visual language model, naming him among its four authors.

  4. December 2023

    Co-lead for multimodal vision

    He is co-lead for multimodal vision on Gemini: A Family of Highly Capable Multimodal Models, submitted to arXiv in December 2023. The work spans image, audio, video and text understanding.

  5. March 2024–July 2025

    Author of three Gemini reports

    He is an author of the Gemini 1.5 report, submitted to arXiv in March 2024, and of the Gemini Robotics report, submitted to arXiv in March 2025. The Gemini 2.5 report, submitted to arXiv in July 2025, also names him as an author.

  6. May 2025–September 2025

    AlphaEvolve and Gemini Robotics 1.5 credits

    The AlphaEvolve post, published in May 2025, thanks him for contributions, advice and support. The Gemini Robotics 1.5 post, published in September 2025, thanks him for leadership and support of this effort.

  7. July 2026

    Leadership thanks on Gemini Robotics 2

    The Gemini Robotics 2 post, published in July 2026, thanks him for leadership and support of this effort.

  8. As of October 2026

    Listed at Google, Mountain View

    A Nature Medicine paper published in October 2026 lists his affiliation as Google in Mountain View.

Sources and credits

Photo: We have no published photo of Jean-Baptiste Alayrac yet. Send us a free-licensed photo.

  1. Jean-Baptiste Alayrac di.ens.fr. Checked 15 October 2026.
  2. [1707.09074] Learning from Video and Text via Large-Scale Discriminative Clustering arxiv.org. Checked 15 October 2026.
  3. [2204.14198] Flamingo: a Visual Language Model for Few-Shot Learning arxiv.org. Checked 15 October 2026.
  4. Tackling multiple tasks with a single visual language model — Google DeepMind deepmind.google. Checked 15 October 2026.
  5. Gemini - A Family of Highly Capable Multimodal Models: Contributions and Acknowledgments | HackerNoon hackernoon.com. Checked 15 October 2026.
  6. [2312.11805] Gemini: A Family of Highly Capable Multimodal Models arxiv.org. Checked 15 October 2026.
  7. [2403.05530] Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context arxiv.org. Checked 15 October 2026.
  8. [2503.20020] Gemini Robotics: Bringing AI into the Physical World arxiv.org. Checked 15 October 2026.
  9. [2507.06261] Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities arxiv.org. Checked 15 October 2026.
  10. AlphaEvolve: A Gemini-powered coding agent for designing advanced algorithms — Google DeepMind deepmind.google. Checked 15 October 2026.
  11. Gemini Robotics 1.5 brings AI agents into the physical world — Google DeepMind deepmind.google. Checked 15 October 2026.
  12. Gemini Robotics 2 brings whole body intelligence to robots — Google DeepMind deepmind.google. Checked 15 October 2026.
  13. An open vision-language model for diverse medical applications | Nature Medicine nature.com. Checked 15 October 2026.