Updated daily at 05:00 UTC
Vision-language
Janus
Small unified models that both understand images and generate them from text.
By DeepSeek.
Versions
4 versions
| Model | Maker | Kind | Released | Context | Input $/1M | Output $/1M |
|---|---|---|---|---|---|---|
| Janus-Pro 1BJanus · Open weights | DeepSeek | Vision-language | 27 Jan 2025 | 16.4K | — | — |
| Janus-Pro 7BJanus · Open weights | DeepSeek | Vision-language | 27 Jan 2025 | 16.4K | — | — |
| JanusFlow 1.3BJanus · Legacy · Open weights | DeepSeek | Vision-language | Nov 2024 | 16.4K | — | — |
| Janus 1.3BJanus · Legacy · Open weights | DeepSeek | Vision-language | 18 Oct 2024 | 16.4K | — | — |
