---
id: "moondream/moondream3-preview"
also: ["moondream/moondream3-preview-20261013", "moondream/moondream3-preview"]
cheapest_list_usd_per_m: "0.3 in, 2.5 out, at primeintellect/fp8 on openrouter"
cheapest_effective_usd_per_m: "0.2983 in, 2.4992 out, at primeintellect/fp8 on openrouter"
context: 32000
max_output: 4096
input: ["text", "image"]
output: ["text"]
released: "2026-10-15"
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["include_reasoning", "max_completion_tokens", "max_tokens", "reasoning", "temperature", "top_p"]
endpoints: 1
gateways: "openrouter (1)"
no_training: 1
trains: 0
retains_prompts: 0
---
# Moondream: Moondream 3 Preview

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...