DeepSeek series comparison
DeepSeek family: open-source and highly cost-effective — the V-series for general use, the R-series for reasoning, at a fraction of closed-flagship pricing.
All models
DeepSeek series comparison
DeepSeek family: open-source and highly cost-effective — the V-series for general use, the R-series for reasoning, at a fraction of closed-flagship pricing.
4 versions (sorted by capability)
| Version | Context | Input | Output | Reasoning | Code | Speed | Open source |
|---|---|---|---|---|---|---|---|
| DeepSeek V4 Flash | 1M | $0.14 / 1M | $0.28 / 1M | ●●●●○ | ●●●●○ | ●●●●● | Open |
| DeepSeek V4 Pro | 1M | $0.44 / 1M | $0.87 / 1M | ●●●●○ | ●●●●● | ●●●○○ | Open |
| DeepSeek-R1 | 64K | $0.55 / 1M | $2.19 / 1M | ●●●●● | ●●●●● | ●●●○○ | Open |
| DeepSeek-V3 | 64K | $0.27 / 1M | $1.1 / 1M | ●●●●● | ●●●●● | ●●●●○ | Open |
DeepSeek V4 FlashDeepSeek's fast open-source model with 1M context and extremely low price, suitable for large-scale batch processing.DeepSeek V4 ProChina's open-source flagship, with 1M context and dual thinking/non-thinking modes, excels in coding and reasoning, priced at a fraction of closed-source flagships.DeepSeek-R1Designed for complex reasoning, with math/logic/coding capabilities comparable to o1, but fully open-source with training costs at only 3%.DeepSeek-V3Domestic flagship model with code and math capabilities comparable to Claude, priced at only 5% of OpenAI, the king of cost-effectiveness.