—
Moondream
Moondream is an AI-native company in —, based in United States. Shown below: 5 models.
US
- Listed models
- 5
Figures and classifications reflect imported catalog records, not live company reporting.
Models
5
-
Photon Gen 3 Multimodal
Moondream’s real-time vision-language model aimed at production video and image analysis. It is designed to deliver VLM-style visual reasoning fast enough for live use cases such as manufacturing inspection, broadcast m…
- Image interpretation
- Visual assistance
- Object identification
- Computer vision assistance
-
Moondream 3 Gen 3 Preview Multimodal
Preview is a compact frontier-oriented vision-language model built for fast visual reasoning, grounding, OCR, object detection, pointing, and structured output. It uses a 9B MoE architecture with 2B active parameters an…
- Image interpretation
- Image recognition
- Image segmentation
- OCR
-
moondream 2 Gen 3
Moondream Moondream 2 is a small open vision-language model designed to run efficiently across many environments. It is the previous-generation Moondream model,
-
Moondream 0.5B Gen 4 Image
A tiny open-source vision-language model built for edge devices and mobile platforms. With only 0.5B parameters, it is positioned as the world’s smallest VLM, designed for fast lightweight deployment on constrained hard…
- Image interpretation
- Image recognition
-
Moondream 1 Gen 3 Multimodal
Moondream Small, efficient open-source vision-language model designed to run broadly on many devices.
- Image interpretation
- Image descriptions
Source rankings
Imported positions; not an independent quality score.
- Overall Rank#2061
- AI Native Rank#1036
- Rank#980