Moondream 3 Preview
by Moondreamtext+image->text1 endpoint
Overview
Moondream 3 Preview is a vision-language model from Moondream with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and visual reasoning over images, with a 32K-token context and an optional reasoning mode.
Capabilities
Text generation
Image understanding
Extended reasoning
Top-P sampling
Modalities
Input
TextImage
Output
Text
Technical Specifications
Context Window
32K tokens
Max Output
4.1K tokens
Tokenizer
Other
Uptime (24h)
10000.0%
Supported Parameters
include_reasoningmax_completion_tokensmax_tokensreasoningtemperaturetop_p
Pricing
Per 1M tokens
Input$0.30
Output$2.50
Added
October 15, 2026
Last synced 10/23/2026