AI People Wiki

Builders & open source

Christopher Clark

Senior research scientist, leading the open-ecosystem multimodal team, Ai2

Christopher Clark is a senior research scientist at the Allen Institute for AI (Ai2), where Clark leads the open-ecosystem multimodal team working on vision-language models with a focus on openness, training efficiency and visual grounding. Clark is co-first author of the Molmo and PixMo paper and first-listed co-first author of the Molmo2 paper, and earlier worked on the Unified-IO line of any-to-any models.1234

At a glance

Current role
Senior research scientist, leading the open-ecosystem multimodal team
Company
Ai2
Sector
Nonprofit
Education
UW
Work listed
3
Honors
1
Sources cited
6
Profile reviewed

Career

Clark received a PhD from the University of Washington, advised by Luke Zettlemoyer, and is a senior research scientist at Ai2. Clark leads Ai2's open-ecosystem multimodal team and has also worked on data generation, generalization, language-model pre-training, game-playing and evaluation.14

Research

Clark's earlier multimodal work centred on unified models: Unified-IO (2022) handled vision, language and multimodal tasks in one model, and Unified-IO 2 (2023) extended this to audio and action. Both papers list Clark as a co-first author.561

Clark's recent work is on the Molmo line of fully open vision-language models, which share weights, code and training data. The Molmo and PixMo paper (2024) received a CVPR 2025 Best Paper Honorable Mention, and the Molmo2 paper (2026), adding video understanding and grounding, was a CVPR 2026 best paper award candidate. Clark is also first author of MolmoPoint, on pointing with grounding tokens, at ECCV 2026.123

Work and contributions

  • Molmo2-O 7B Open weights

    First-listed co-first author of the Molmo2 paper, which describes the open-weight Molmo2 vision-language models with video understanding and grounding.31

  • Molmo Open weights

    Co-first author, with Matt Deitke, of the Molmo and PixMo paper introducing the Molmo family of open vision-language models and the PixMo datasets.21

  • Unified-IO and Unified-IO 2 Published research

    Co-first author of Unified-IO (2022) and Unified-IO 2 (2023), unified models spanning vision, language, audio and action.561

Timeline

  1. Posted Unified-IO, a unified model for vision, language and multimodal tasks, as co-first author.5
  2. Posted Unified-IO 2, adding audio and action, as co-first author.6
  3. Posted the Molmo and PixMo paper as co-first author with Matt Deitke.2
  4. Received a CVPR 2025 Best Paper Honorable Mention for the Molmo and PixMo paper.1
  5. First author of MolmoPoint, on pointing with grounding tokens, listed for ECCV 2026.1
  6. Posted the Molmo2 paper on open vision-language models with video understanding as first-listed author.3

Honors and recognition

  • CVPR Best Paper Honorable Mention (Molmo and PixMo) · IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 20251

Where they work

  • Ai2 23 people in the wiki

Suggest a correction or removal

Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.

References

6 sources · profile reviewed Oct 22, 2026 · data updated Oct 22, 2026

  1. Christopher Clark — homepage and publicationsChristopher Clark · Official page · accessed Oct 22, 2026
  2. Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models (arXiv 2409.17146)arXiv · Paper · published Sep 25, 2024 · accessed Oct 22, 2026
  3. Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Grounding (arXiv 2601.10611)arXiv · Paper · published Jan 15, 2026 · accessed Oct 22, 2026
  4. Ai2 — Team directoryAllen Institute for AI · Official page · accessed Oct 22, 2026
  5. Unified-IO: A Unified Model for Vision, Language, and Multi-Modal Tasks (arXiv 2206.08916)arXiv · Paper · published Jun 17, 2022 · accessed Oct 22, 2026
  6. Unified-IO 2: Scaling Autoregressive Multimodal Models with Vision, Language, Audio, and Action (arXiv 2312.17172)arXiv · Paper · published Dec 28, 2023 · accessed Oct 22, 2026