Abstract

We introduce HyperBrowseComp, a multilingual and multimodal browsing benchmark comprising 423 manually authored and human-validated questions across 13 languages, written by native or highly proficient speakers. Questions are designed to be extremely challenging. Each question targets a concise, publicly verifiable answer whose discovery requires locating obscure evidence, following multi-step clue chains, or inspecting heterogeneous sources such as videos, scanned documents, images, or maps. Easier questions are filtered out by evaluating them with models without internet access to reduce the likelihood that they can be answered with parametric knowledge alone. We evaluate several models using provider-native search and a shared external retrieval harness under a common agent protocol. To contextualize model performance and effort, we also conduct a human evaluation on a sample of the questions. HyperBrowseComp provides a challenging testbed for persistent information seeking across languages and evidence modalities, with difficulty arising from discovering and connecting evidence on the open web.

Keywords

Subject

Publication details

Journal
Not available
Open access
Green open access

Cite this article

APA 7

Aji, A. F., Ramadhan, F. R., Zuhri, Z. M. K., Han, S. H. E., Diandaru, R., Cui, Q., Cruz, J. C. B., Chandana, B., Chomphooyod, P., Attia, A., Mansurov, J., Villa-Cueva, E., Nguyen, C. D., Turganov, I., Wu, M., Limkonchotiwat, P., & Nikishina, I. (2026). HyperBrowseComp: A Multilingual and Multimodal Stress Test for Web-Browsing Agents. https://omanscience.com/en/articles/hyperbrowsecomp-a-multilingual-and-multimodal-stress-test-for-web-browsing-agents

MLA 9

Aji, Alham Fikri, et al. "HyperBrowseComp: A Multilingual and Multimodal Stress Test for Web-Browsing Agents." https://omanscience.com/en/articles/hyperbrowsecomp-a-multilingual-and-multimodal-stress-test-for-web-browsing-agents.

Chicago (author–date)

Aji, Alham Fikri, Faiz Rizki Ramadhan, Zayd M. K. Zuhri, Seung Hun Eddie Han, Ryandito Diandaru, Qinrong Cui, Jan Christian Blaise Cruz, Badrinath Chandana, Peerawat Chomphooyod, Ahmed Attia, Jonibek Mansurov, Emilio Villa-Cueva, Canh Duong Nguyen, Imran Turganov, Minghao Wu, Peerat Limkonchotiwat, and Irina Nikishina. 2026. "HyperBrowseComp: A Multilingual and Multimodal Stress Test for Web-Browsing Agents." https://omanscience.com/en/articles/hyperbrowsecomp-a-multilingual-and-multimodal-stress-test-for-web-browsing-agents.

Harvard

Aji, A. F., Ramadhan, F. R., Zuhri, Z. M. K., Han, S. H. E., Diandaru, R., Cui, Q., Cruz, J. C. B., Chandana, B., Chomphooyod, P., Attia, A., Mansurov, J., Villa-Cueva, E., Nguyen, C. D., Turganov, I., Wu, M., Limkonchotiwat, P. and Nikishina, I. (2026) 'HyperBrowseComp: A Multilingual and Multimodal Stress Test for Web-Browsing Agents', Available at: https://omanscience.com/en/articles/hyperbrowsecomp-a-multilingual-and-multimodal-stress-test-for-web-browsing-agents.

Vancouver

Aji AF, Ramadhan FR, Zuhri ZMK, Han SHE, Diandaru R, Cui Q, et al. HyperBrowseComp: A Multilingual and Multimodal Stress Test for Web-Browsing Agents. https://omanscience.com/en/articles/hyperbrowsecomp-a-multilingual-and-multimodal-stress-test-for-web-browsing-agents

IEEE

A. F. Aji, F. R. Ramadhan, Z. M. K. Zuhri, S. H. E. Han, R. Diandaru, Q. Cui, J. C. B. Cruz, B. Chandana, P. Chomphooyod, A. Attia, J. Mansurov, E. Villa-Cueva, C. D. Nguyen, I. Turganov, M. Wu, P. Limkonchotiwat, and I. Nikishina, "HyperBrowseComp: A Multilingual and Multimodal Stress Test for Web-Browsing Agents," https://omanscience.com/en/articles/hyperbrowsecomp-a-multilingual-and-multimodal-stress-test-for-web-browsing-agents.