Moondream
Commercial provider · 2 models tracked
The maker of Moondream, a family of small vision language models built by M87 Labs. The current Moondream 3 line is a mixture-of-experts with 9B total and 2B active parameters that answers questions about images, captions, reads text and points at, counts and detects objects. Moondream publishes the weights on Hugging Face, sells hosted inference per token on Moondream Cloud, and offers fine-tuning through Lens. Moondream states no founding year on its own pages, so none is published here.
Moondream pricing at a glanceOctober 2026 · $ per 1M tokens
Moondream API pricing (October 2026): 2 current models range from $0.300 to $0.300 per 1M input tokens and $1.00 to $2.50 per 1M output tokens. The cheapest paid model is Moondream 3.1 9B A2B at $0.300/1M input; the priciest is Moondream 3 Preview at $0.300/1M input / $2.50 output. Every price links to Moondream's official pricing page and refreshes twice daily.
Today's Moondream prices
2 models · cheapest blended firstSorted by blended cost (cheapest first). Prices per 1M tokens, October 2026 — every price links to Moondream's official pricing page.
| Model | Blended* | Input | Output | Cached in | Relative cost | Context | Status |
|---|---|---|---|---|---|---|---|
| Moondream 3.1 9B A2B | $0.475 |
$0.300 | $1.00 | — | 32K | Current | |
| Moondream 3 Preview | $0.850 |
$0.300 | $2.50 | — | 32K | Current |
Quick stats
- Models tracked
- 2
- Type
- commercial
- Founded
- —
- Cheapest model
- Moondream 3.1 9B A2B
- Cheapest blended
- $0.475/M
Open-weight models
2 open-weight models available from this provider.
Self-host Moondream's open-weight models
These models ship with open weights, so you can serve them yourself on rented GPUs instead of paying per-token API prices — often cheaper at scale.
Partner links — we may earn a commission if you sign up, at no cost to you. We only list platforms we'd recommend regardless.