DeepSeek models
11models tracked · ranked by intelligence, speed & price
Match · Updated July 2026
DeepSeek's smartest model is DeepSeek V4 Pro (44.3 on the Intelligence Index), and its cheapest is DeepSeek V4 Flash at $0.140 per million input tokens. All 11 DeepSeek models are compared below by intelligence, speed, latency, context and price.
Smartest DeepSeekFastest DeepSeekCheapest DeepSeekSmall & fast DeepSeekDeepSeek for codingDeepSeek reasoningDeepSeek for designDeepSeek for agents
Smartest
44.3
deepseek/deepseek-v4-pro
Fastest
—
Cheapest
$0.140/M
deepseek/deepseek-v4-flash
| Model | Intel | Speed | Latency | In $/M | Context |
|---|---|---|---|---|---|
| deepseek-v4-pro ReasoningToolsJSON | 44.3 | — | — | $0.435 | 1.0M |
| deepseek-v4-flash ReasoningToolsJSON | 40.3 | — | — | $0.140 | 1.0M |
| deepseek-v3.2 ReasoningToolsJSON | 24.7 | — | — | $0.269 | 164K |
| deepseek-v3.1-terminus ReasoningToolsJSON | 21.4 | — | — | $0.270 | 164K |
| deepseek-r1 ReasoningToolsJSON | 20.1 | — | — | $0.700 | 164K |
| deepseek-r1-distill-llama-70b Reasoning | 9.9 | — | — | $0.800 | 8K |
| deepseek-v3.2-exp ReasoningToolsJSON | — | — | — | $0.270 | 164K |
| deepseek-chat-v3.1 ReasoningToolsJSON | — | — | — | $0.250 | 164K |
| deepseek-r1-0528 ReasoningToolsJSON | — | — | — | $0.500 | 164K |
| deepseek-chat-v3-0324 ToolsJSON | — | — | — | $0.270 | 164K |
| deepseek-chat ToolsJSON | — | — | — | $0.257 | 164K |