Moondream 3.1
Moondream's vision-language model (9B total, 2B active parameters) for questions about images, captions, reading text in images and visual reasoning.
Moondream's Moondream 3.1 is a cloud text model. It is multimodal: alongside a text prompt it accepts images, then replies with generated text. Its context window handles up to 32K input tokens and up to 4K output tokens. Generation can be tuned with system instructions, temperature, and top-p sampling. It runs through OpenRouter using your own API key, from $0.30 per million input tokens.
- Released
- Oct 2026
- Pricing
- $0.30 / 1M in · $2.50 / 1M out
- Inputs
- Text, Images
- Context window
- 32K in · 4K out
- Controls
- System prompt, Temperature, Top-p
Samples
Generated with Moondream 3.1 via OpenRouter. The same three prompts run against every text model in the catalog, shown verbatim, so the only thing that changes between two models’ answers is the model.
Explain what an API is to someone who has never written code. Do not use a restaurant or waiter analogy. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 50 words.
An API, or Application Programming Interface, is a set of rules that lets different software systems talk to each other. It acts as an intermediary, enabling one program to request services from another without needing to know how they work internally. APIs allow data exchange, integration, and automation across applications and platforms.
A two-person startup is deciding whether to build their own user authentication or use a hosted provider. Give them a clear recommendation, then the single strongest argument against your own recommendation. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 150 words.
Build your own authentication: Strong advantage over third-party providers. Direct control over the user experience, security posture, and data handling. Cost-effective for small teams with limited budgets. Customization options to fit specific needs. Better data ownership and compliance with privacy regulations. Community-driven development for continuous improvement. Strongest argument: cost-effectiveness and direct control over the authentication process outweigh the perceived benefits of using a hosted provider. Strongest argument: cost-effectiveness and direct control over the authentication process outweigh the perceived benefits of using a hosted provider.
Write a briefing for a non-technical executive explaining why their company's spending on AI is hard to predict, and what they can do about it. Cover at least three distinct causes. Plain prose only — no headings, bullet points, or markdown formatting. Write exactly 205 words.
AI spending is challenging to predict because companies are investing in technologies that evolve rapidly. Firstly, AI models require continuous learning from vast datasets, which demands significant computational resources and expertise. Secondly, ethical considerations around bias and privacy in AI systems require careful oversight and adaptation. Thirdly, market dynamics and competitive pressures influence investment decisions, making it difficult to forecast long-term ROI. These factors mean AI spending requires agility and adaptability rather than static budgeting, making it harder to predict than traditional technology investments.
Moondream
Moondream builds small, efficient vision-language models for image questions, captions and reading text in images.
moondream.ai ↗