DeepSeek-V3.1 is post-trained on the top of DeepSeek-V3.1-Base, which is built upon the original V3 base checkpoint through a two-phase long context extension approach, following the methodology outlined in the original DeepSeek-V3 report. We have expanded our dataset by collecting additional long documents and substantially extending both training phases. The 32K extension phase has been increased 10-fold to 630B tokens, while the 128K extension phase has been extended by 3.3x to 209B tokens...
function_callinghttps://api.deepinfra.com/v1/openai/v1/chat/completions| Model | Context | Max out | Pricing |
|---|---|---|---|
| DeepSeek-V3-0324 | 164K | — | $0.24$0.9 |
| DeepSeek-V3.1 | 164K | — | $0.25$0.95 |
| DeepSeek-V3.2 | 164K | — | $0.26$0.38 |
| DeepSeek-V3.1-Terminus | 164K | — | $0.27$0.95 |
| DeepSeek-V3 | 164K | — | $0.32$0.89 |
/v1/models/deepinfra/deepseek-ai/DeepSeek-V3.1DeepSeek-V3.1 is post-trained on the top of DeepSeek-V3.1-Base, which is built upon the original V3 base checkpoint through a two-phase long context extension approach, following the methodology outlined in the original DeepSeek-V3 report. We have expanded our dataset by collecting additional long documents and substantially extending both training phases. The 32K extension phase has been increased 10-fold to 630B tokens, while the 128K extension phase has been extended by 3.3x to 209B tokens...