Molmo 2 | Counting objects and actions @allenai

Uploaded December 2025 | Updated September 2026, 4 hours ago

Researcher Zixin Ma demonstrates Molmo 2's counting capabilities.

Molmo 2 is a state-of-the-art open multimodal model suite capable of precise spatial and temporal understanding of video, image, and multi-image sets. Building on the global impact of Molmo, which pioneered image pointing for multimodal AI systems, Molmo 2 introduces breakthrough capabilities in video pointing, multi-frame reasoning, and object tracking.

Learn more: allenai.org/molmo
Discuss in Discord: discord.gg/ai2

BLADE: Benchmarking Language Model Agents for Data-Driven Science

Just-DREAM-about-it: Figurative Language Understanding with DREAM-FLUTE

Helping or Herding? Reward Model Ensembles Mitigate but do not Eliminate Reward Hacking

Towards the Age of Computation and AI for High Performance Climate

When Not to Trust Language Models: Investigating Effectiveness of Parametric&Non-Parametric Memories

Open-Ended Learning Leads to Generally Capable Agents | Embodied AI Lecture Series at AI2

From LLMs to Agents: Generalizability from the Inside Out

Data-Centric Approaches to Adapting Foundation Models

The University of Washington eScience Institute: a Home for Data-Intensive Discovery

Generalization for Robot Learning In The Wild | Embodied AI Lecture series at AI2

Towards Generalist Agents for Accelerating Scientific Discovery171