DeepSeek V3.1
LLM
by deepseek ·
deepseek-deepseek-chat-v31
Context window
164K
Max output
33K
Providers
9
Best input
$0.25
About this model
DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context...
Providers & prices
| Provider | Input $/MTok | Output $/MTok | Context | Upstream ID |
|---|---|---|---|---|
| D DeepInfra | $0.25 | $0.95 | 164K | deepseek/deepseek-chat-v3.1 |
| D DeepSeek | $0.25 | $0.95 | 164K | deepseek/deepseek-chat-v3.1 |
| NA Novita AI | $0.27 | $1.00 | 131K | deepseek/deepseek-chat-v3.1 |
| S SiliconFlow | $0.27 | $1.00 | 164K | deepseek/deepseek-chat-v3.1 |
| A AtlasCloud | $0.3 | $0.95 | 131K | deepseek/deepseek-chat-v3.1 |
| C CoreWeave | $0.55 | $1.65 | 161K | deepseek/deepseek-chat-v3.1 |
| G Google | $0.6 | $1.70 | 164K | deepseek/deepseek-chat-v3.1 |
| M Mara | $0.6 | $1.70 | 131K | deepseek/deepseek-chat-v3.1 |
| S SambaNova | $0.65 | $1.50 | 131K | deepseek/deepseek-chat-v3.1 |