Moondream 3.1
Moondream 3.1 is a large language model made by Moondream, added to the catalog on 15 Oct 2026. It has a context window of 32,000 tokens and costs $0.30 per 1M input tokens and $2.50 per 1M output tokens. It accepts text and images and produces text. Typical uses: answering questions about images, writing image captions and reading text in images.
Price checked ยท source: OpenRouter
What Moondream 3.1 is
Moondream 3.1 is a vision-language model from Moondream and the successor to Moondream 3 Preview. It accepts text and images and produces text. Its listed visual tasks include visual question answering, captioning, and optical character recognition (OCR), so it can be used to ask questions about image content, describe images in text, or read text present in images. The model also lists reasoning support. The catalog identifies it as a mixture-of-experts model with 9 billion total parameters and 2 billion active parameters.
The model has a context length of 32,000 tokens and a maximum output of 4,096 tokens. That context allows a prompt to include text alongside image-based requests, within the stated limit, and gives room for responses to visual questions or image descriptions. Its text output suits workflows that need written answers or extracted text rather than generated images. The available details establish image and text input, text output, reasoning support, and the named vision tasks; they do not specify additional input or output modalities. The stated maximum output is shorter than its context length, and both limits are given in tokens. Moondream 3.1 may suit tasks involving images and language, including extracting visible text, producing image captions, and responding to questions about supplied visuals.
Key features
- Accepts text and image inputs
- Produces text output
- Supports reasoning
- Has a 32,000-token context length
- Supports visual question answering, captioning, and OCR
Typical uses
- Answering questions about images
- Writing image captions
- Reading text in images
- Extracting visible text from images
Moondream 3.1 pricing
US dollars, list price through OpenRouter.
| Item | Standard |
|---|---|
| Input, per 1M tokens | $0.30 |
| Output, per 1M tokens | $2.50 |
No price change since it was added on 15 Oct 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0009 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0095 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.014 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.049 |
Moondream 3.1 specifications
| Model id | moondream/moondream3.1-9b-a2b |
|---|---|
| Version | moondream3.1-9b-a2b-20261013 |
| Context window | 32,000 tokensAbout 24,000 English words |
| Max output | 4,096 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 15 Oct 2026 |
| Input types | text, image |
|---|---|
| Output types | text |
| Tool calling | No |
| Structured outputs | No |
| Reasoning effort control | No |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for Moondream 3.1 yet.
Compare Moondream 3.1
Side by side with other models: prices, context window, what each accepts and supports, and the same four tasks priced on both.