OpenAI: GPT-5.1-Codex wins on more metrics (5 of 9), but the right pick depends on what you optimize for — see the breakdown below.
| Metric | Anthropic: Claude Sonnet 4.6 | OpenAI: GPT-5.1-Codex |
|---|---|---|
| Intelligence Index | 42.6 | 43.1✓ |
| Coding Index | 43.0✓ | 36.6 |
| GPQA Diamond | 80% | 86%✓ |
| Design Arena Elo | 1328✓ | 1218 |
| Speed (tokens/sec) | 46 | 60✓ |
| Latency | 1.1s✓ | 2.8s |
| Input price /M | $3.00 | $1.25✓ |
| Output price /M | $15.00 | $10.00✓ |
| Context window | 1M✓ | 400K |
| Capabilities | ReasoningToolsJSONVision | ReasoningToolsJSONVision |