PeopleofSuperintelligence

Researcher

Rahim Entezari

Rahim Entezari is a deep learning researcher who worked on Stable Diffusion 3 at Stability AI. During a PhD at Graz University of Technology and the Complexity Science Hub in Vienna, he coauthored a 2021 conjecture on permutation invariance and linear mode connectivity of neural networks, and he earned the degree with distinction in 2023. After joining Stability AI as a research scientist in May 2023, he became one of the 17 authors of Scaling Rectified Flow Transformers for High-Resolution Image Synthesis, the Stable Diffusion 3 paper, which won a Best Paper award at ICML 2024. In 2025 he coauthored SD3.5-Flash, a few-step distillation framework for image generation on consumer devices, and Stable Cinemetrics, a benchmark for professional video generation. He is a senior research scientist at Wayve.

Now at
Wayve
Badges

Timeline

  1. 2019–2023

    PhD at TU Graz

    He becomes a PhD student at the Institute for Technical Informatics at Graz University of Technology and at the Complexity Science Hub in Vienna in 2019. In September 2019, he coauthors Class-dependent Compression of Deep Neural Networks with Olga Saukh, an iterative technique whose compressed networks use up to 99% fewer parameters.

  2. 2021–2023

    Permutation invariance and loss barriers

    In 2021, he coauthors a paper with Hanie Sedghi, Olga Saukh and Behnam Neyshabur that conjectures SGD solutions likely have no barrier in the linear interpolation between them once the permutation invariance of neural networks is taken into account. REPAIR, an ICLR 2023 paper he coauthors, follows up on the conjecture and reports 60%–100% relative barrier reduction across architecture families and tasks.

  3. 2022–2023

    DataComp and a Washington internship

    He interns at the University of Washington in 2022. In 2023, he is among the authors of DataComp, a testbed for dataset experiments built around a pool of 12.8 billion image-text pairs from Common Crawl.

  4. 2023

    A PhD and Stability AI

    He joins Stability AI as a research scientist in May 2023 and earns his PhD with distinction in July 2023. At Stability, he works on text-to-image and text-to-video models.

  5. March 2024

    The Stable Diffusion 3 paper

    In March 2024, he is among the 17 authors of Scaling Rectified Flow Transformers for High-Resolution Image Synthesis, the Stable Diffusion 3 paper. It improves noise sampling for training rectified flow models and presents a transformer architecture with separate weights for the image and text modalities.

  6. 2024

    A Best Paper at ICML

    Scaling Rectified Flow Transformers for High-Resolution Image Synthesis receives a Best Paper award at ICML 2024, with him among its authors.

  7. 2025

    SD3.5-Flash and Stable Cinemetrics

    In September 2025, he coauthors SD3.5-Flash, a few-step distillation framework that brings image generation to consumer devices from mobile phones to desktop computers. He also coauthors Stable Cinemetrics, an evaluation framework for professional video generation that is accepted at NeurIPS 2025.

  8. As of October 2026

    Senior researcher at Wayve

    He is a senior research scientist at Wayve, after joining Stability AI as a researcher in 2023.

Career

  1. NowWayveSenior research scientist
  2. From 2023Stability AIResearch scientist
  3. 2022University of WashingtonIntern

Education

  1. 2023Graz University of TechnologyPhD
  2. 2023Complexity Science HubPhD

From the biography.

  1. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes University of Washington. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  2. The median h-index is 48 at universities and 14 at frontier AI labsIncludes University of Washington. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  3. Of 943 people shaping AI who worked at Google, 446 work there nowIncludes University of Washington. Our estimates of where the 943 people shaping AI who worked at Google and Google DeepMind work now: 446 at Google and 452 across 278 other organizations.
  4. 20% of people at companies who shape AI also hold a university role, led by Google and Stanford UniversityIncludes University of Washington. We estimate that 539 of the 6,766 people with a known current organization (8%) hold both a university role and a company role, including 20% of people at companies and 13% of people at universities.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Rahim Entezari yet. Send us a free-licensed photo.

  1. Rahim Entezari rahimentezari.github.io. Checked 22 October 2026.
  2. Class-dependent Compression of Deep Neural Networks arxiv.org. Checked 22 October 2026.
  3. Rahim Entezari — Lacuna lacuna.tiptreesystems.com. Checked 22 October 2026.
  4. The Role of Permutation Invariance in Linear Mode Connectivity of Neural Networks arxiv.org. Checked 22 October 2026.
  5. REPAIR: REnormalizing Permuted Activations for Interpolation Repair arxiv.org. Checked 22 October 2026.
  6. DataComp: In search of the next generation of multimodal datasets arxiv.org. Checked 22 October 2026.
  7. Scaling Rectified Flow Transformers for High-Resolution Image Synthesis arxiv.org. Checked 22 October 2026.
  8. ICML 2024 Awards icml.cc. Checked 22 October 2026.
  9. SD3.5-Flash: Distribution-Guided Distillation of Generative Flows arxiv.org. Checked 22 October 2026.
  10. Stable Cinemetrics: Structured Taxonomy and Evaluation for Professional Video Generation arxiv.org. Checked 22 October 2026.
  11. Rahim Entezari - Google Scholar scholar.google.com. Checked 22 October 2026.