DeepSeek: DeepSeek V4 Flash vs OpenAI: GPT-5.2-Codex
DeepSeek: DeepSeek V4 Flash wins on more metrics (8 of 9), but the right pick depends on what you optimize for — see the breakdown below.
| Metric | DeepSeek: DeepSeek V4 Flash | OpenAI: GPT-5.2-Codex |
|---|---|---|
| Intelligence Index | 40.3✓ | 40.1 |
| Coding Index | 56.2✓ | — |
| GPQA Diamond | 89% | 90%✓ |
| Design Arena Elo | 1268✓ | — |
| Speed (tokens/sec) | 89✓ | 76 |
| Latency | 570ms✓ | 2.2s |
| Input price /M | $0.090✓ | $1.75 |
| Output price /M | $0.180✓ | $14.00 |
| Context window | 1.0M✓ | 400K |
| Capabilities | ReasoningToolsJSON | ReasoningToolsJSONVision |