Intelligence Index
0153.2
> 90% OF MODELS
DeepSeek
DeepSeek V4 Pro
Updated: August 2026
DeepSeek's flagship 1M-context MoE model for advanced reasoning, full-codebase analysis, and long-horizon agent workflows.
Specifications
- Context
- 1M
- Input Modalities
- Text, File
- Output Modalities
- Text
- Input
- $0.435/M
- Output
- $0.87/M
Capabilities
THINKINGCODINGWEB
Similarly Priced Models
| Model | Provider | Context | Input Price | Output Price |
|---|---|---|---|---|
| Qwen3.8 27B | Qwen | 262K | $0.45/M | $3.2/M |
| Kimi K2.5 | Moonshot | 262K | $0.45/M | $2.2/M |
| Qwen3 235B A22B | Qwen | 131K | $0.455/M | $1.82/M |
| Qwen3.7 Plus | Qwen | 1M | $0.4/M | $1.6/M |
| MiMo V2.5 |  Xiaomi | 1M | $0.4/M | $2/M |
Performance Metrics
Coding Index
0268.8
> 79% OF MODELS
Agentic Index
0349.6
> 90% OF MODELS
Average Response Performance
- Output Speed
- 74.2 tok/s
- Time To First Token
- 1.7s
- Time To First Answer Token
- 28.65s
- End To End Response Time
- 35.39s
DATA SOURCE: Artificial Analysis