DeepSeek API Pricing
2 models · From $0.150/1M to $0.660/1M input
The cheapest DeepSeek model to run is DeepSeek V4.1 Flash at $0.150 input and $0.600 output per 1M tokens. On a RAG answer of 8,000 tokens in and 600 out, DeepSeek models range from $0.0016 (DeepSeek V4.1 Flash) to $0.0065 (DeepSeek V4-Pro) per request. The largest context window is DeepSeek V4.1 Flash at 1.0M tokens. Count tokens with the DeepSeek token counter.
| Model | Input /1M | Output /1M | RAG answer |
|---|---|---|---|
| DeepSeek V4.1 Flash | $0.150 | $0.600 | $0.0016 |
| DeepSeek V4-Pro | $0.660 | $1.98 | $0.0065 |