DEEP REASONING · 高嶺
DeepSeek
3 models from DeepSeek, living in The High Peaks.
Visit the DeepSeek village in 3D
| Model | Context | Input | Output | Accepts |
|---|---|---|---|---|
| DeepSeek V4 Pro 0813deepseek/deepseek-v4-pro-0813 | 1.05M | $0.660 | $1.98 | text |
| DeepSeek V4 Pro 0423deepseek/deepseek-v4-pro | 1.05M | $0.260 | $0.510 | text |
| R1 0528deepseek/deepseek-r1-0528 | 164K | $0.500 | $2.15 | text |
Questions and answers
Which DeepSeek model has the largest context window?
DeepSeek V4 Pro 0423 (DeepSeek), with 1.05M tokens.
Which DeepSeek model is cheapest per input token?
DeepSeek V4 Pro 0423, at $0.260 per million input tokens ($0.510 per million output tokens).
Which DeepSeek models accept images?
None in the 2026-10-24 snapshot.
Which DeepSeek models support reasoning?
DeepSeek V4 Pro 0813, DeepSeek V4 Pro 0423, R1 0528 (3 of 3).
Answers are computed from the 2026-10-24 catalog snapshot.