Molmo 7B-D
by Allen Institute for AI
· MOLMO family · released Sep 26, 2024
Open weights
Active
Fully open PointData vision model.
This model has no first-party API. Open-weight release: self-host from published weights.
Specifications
| Model type | Multimodal |
|---|---|
| Architecture | Dense |
| Parameters | 7 |
| Context window | 32.8K tokens |
| Max output | Not disclosed tokens |
| Knowledge cutoff | Not disclosed |
| License | Apache License 2.0 |
| Modalities | Text input Image input Text output |
| Tags | Small LM Vision |
| End of life | Not disclosed |
| Last verified | Aug 20, 2026 |