Step 5 Preview scores 54.9 on the Respan Index (#12), ahead of GPT-5.6 Terra at 53.5 (#14). They split the 4 benchmarks both report evenly. Step 5 Preview is 3.2x cheaper per token at list price. GPT-5.6 Terra has the larger context window (1.1M tokens).
| Benchmark | GPT-5.6 Terra | Step 5 Preview |
|---|---|---|
| Terminal-Bench 4.0 (Vals AI) | 22.7% | 31.8% |
| LMArena Agent (LMArena) | 0.004 | 0.0048 |
| LMArena Elo | 1466 | 1454 |
| LMArena Vision (LMArena) | 1268 | 1260 |
Higher is better. Each score links to where it was published.
* Estimated: no published scores in that category. How the Respan Index works