Moondream 3 Preview vs Moondream 3.1
Moondream 3 Preview (Moondream) and Moondream 3.1 (Moondream) have the same list price: $0.30 per 1M input tokens and $2.50 per 1M output tokens. Both hold 32,000 tokens of context. Both accept text and images and produce text.
Moondream 3 Preview and Moondream 3.1 both accept text and images, produce text, support reasoning, and have a 32,000-token context and 4,096-token output limit. Moondream 3.1 is described as the successor to Moondream 3 Preview.
Compared from catalog data checked
Side by side
0.00 the better figure in a row| Specification | Moondream 3 Preview | Moondream 3.1 |
|---|---|---|
| Maker | Moondream | Moondream |
| Added to catalog | 15 Oct 2026 | 15 Oct 2026 |
| Input, per 1M tokens | $0.30 | $0.30 |
| Output, per 1M tokens | $2.50 | $2.50 |
| Context windowTokens | 32,000 | 32,000 |
| Max outputTokens | 4,096 | 4,096 |
| Accepts | image, text | image, text |
| Produces | text | text |
| Tool calling | No | No |
| Structured outputs | No | No |
| Reasoning effort control | No | No |
List prices through OpenRouter, in US dollars. Each model's page has its variants, price history and sample outputs.
What the same task costs
List price for one request, from the token counts shown. The same four tasks are priced on every model page.
| Task | Moondream 3 Preview | Moondream 3.1 |
|---|---|---|
| Answer a short question500 in / 300 out | $0.0009 | $0.0009 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0095 | $0.0095 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.014 | $0.014 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.049 | $0.049 |
Which to choose
Choose Moondream 3 Preview if
No row of the table favours Moondream 3 Preview. The difference, if there is one, is in the answers it writes: try it on your own task.
Moondream 3 Preview detailsChoose Moondream 3.1 if
No row of the table favours Moondream 3.1. The difference, if there is one, is in the answers it writes: try it on your own task.
Moondream 3.1 detailsNeither maker's quality claims are repeated here. The reliable test is your own task: ask both the same question and read the two answers.