Anthropic: Claude Opus 4.5 wins on more metrics (5 of 9), but the right pick depends on what you optimize for — see the breakdown below.
| Metric | Anthropic: Claude Opus 4.5 | StepFun: Step 3.7 Flash |
|---|---|---|
| Intelligence Index | 43.1✓ | 42.6 |
| Coding Index | 42.9✓ | 37.1 |
| GPQA Diamond | 81%✓ | 81% |
| Design Arena Elo | 1297✓ | 1232 |
| Speed (tokens/sec) | 58 | 74✓ |
| Latency | 772ms✓ | 1.8s |
| Input price /M | $5.00 | $0.200✓ |
| Output price /M | $25.00 | $1.15✓ |
| Context window | 200K | 256K✓ |
| Capabilities | ReasoningToolsJSONVision | ReasoningToolsJSONVision |