Back to models
Moondream 3.1 9B-A2B
MoEotherReleased Jul 7, 2026
Moondream's vision-language mixture-of-experts model with 9B total and 2B active parameters for visual question answering, captioning, object detection, and pointing. Released July 7, 2026 under the Moondream Model License 1.0, with local inference through Photon and hosted access. The listed 32,000-token context reflects the OpenRouter endpoint.
ChatText completionVisionReasoningtext · image
Provider (1)
Sorted by total cost (input + output per 1M tokens). Select a row to view provider details.
| Provider | Pricing (per 1M) | Rate limits | Regions | Health | Latency |
|---|---|---|---|---|---|
 | In: $0.30Out: $2.50 | N/A RPM / N/A TPM | us-east-1 | Healthy | 0ms |
Quick start
Use this model via OpenRouter with an OpenAI-compatible SDK.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://openrouter.ai/api/v1",
apiKey: process.env.OPENROUTER_API_KEY,
});
const response = await client.chat.completions.create({
model: "moondream/moondream3.1-9b-a2b",
messages: [
{ role: "user", content: "Hello!" }
],
});
console.log(response.choices[0].message.content);Using OpenRouter API · OpenAI-compatible SDK