Molmo 2 8B

Open weights Active Multimodal Dense molmo-2-8b

Open vision-language model on a Qwen3 8B backbone for image and video understanding, pointing and tracking.

Released
Dec 11, 2025
Context window
Not disclosed
Max output
Not disclosed
Parameters
8.7B
Input price
Not disclosed
Output price
Not disclosed

Prices are first-party list prices in USD at the standard tier. Last verified Oct 30, 2026.

Pricing

No first-party API. This is an open-weights release you can run yourself.

Specifications

Developer
Allen Institute for AI
Model type
Multimodal
Architecture
Dense
Parameters
8.7B
Input
Text Image Video
Output
Text
Knowledge cutoff
Not disclosed
License
Apache License 2.0
API access
No first-party API
Capabilities
Vision