Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use
Specification
most-agreed values
Context
128K
Max output
32K
Released
2025-12-01
Knowledge cutoff
2024-07
Retires
—
Open weights
yes
Input
text
Output
text
Price
US dollars per million tokens · most-agreed
Input
$0.2899
Output
$0.4348
Cache read
$0.029
Available from 2 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| volcengine | free | free | — | — | 98K | 33K | — |
| qiniu-ai | $0.2899 | $0.4348 | $0.029 | — | 128K | 32K | — |
More from DeepSeek
most-hosted first