GPT-5.5 Pro vs Step 5 Preview
GPT-5.5 Pro 
Step 5 Preview Verdict
- GPT-5.5 Pro (xhigh) leads on quality: 67.1 vs 65.0.
- GPT-5.5 Pro (xhigh) is stronger in maths.
- Step 5 Preview is stronger in reasoning, composite, knowledge, long context, multimodal.
- Step 5 Preview is 47× cheaper ($1.43 vs $67.50 per 1M blended).
| Metric | GPT-5.5 Pro (xhigh) | Step 5 Preview |
|---|---|---|
| BenchLeader Index | 67.1 | 65.0 |
| Maths score | 74.5 | – |
| Reasoning score | 80.1 | 82.1 |
| Composite score | – | 84.4 |
| Knowledge score | – | 72.0 |
| Long context score | – | 70.8 |
| Multimodal score | – | 60.8 |
| Blended price $/M | $67.50 | $1.43 |
| Output speed | 4 tok/s | – |
| Time to first answer | 3.5 s | – |
| Context window | 1.1M | 1M |
| FrontierMath Tiers 1–3 | 87.7% | – |
| FrontierMath Tier 4 | 78.0% | – |
| AA Intelligence Index | – | 43.6 |
| AA-LCR | – | 88.3% |
| MMMU-Pro | – | 76.4% |
| AA-Omniscience | – | 16.4 |
| Humanity's Last Exam (AA) | – | 46.5% |
| SciCode (AA) | – | 58.9% |
| ARC-AGI-1 | 95.0% | – |
| ARC-AGI-2 | 84.2% | – |
| CritPt | 30.6% | 20.9% |
| GDPval (AA) | – | 53.6% |
| LMCA | 53.9% | – |
| DTBench | 96.0% | – |
Data as of 2026-09-19. Best configuration of each model; every score links to its source on the model pages.
GPT-5.5 Pro vs Step 5 Preview: questions
- Is GPT-5.5 Pro better than Step 5 Preview?
- GPT-5.5 Pro (xhigh) leads on quality: 67.1 vs 65.0. The BenchLeader Index combines every independent quality benchmark; GPT-5.5 Pro (xhigh) is ahead overall as of 2026-09-19, but check the category scores for your use.
- Which is cheaper, GPT-5.5 Pro or Step 5 Preview?
- Step 5 Preview is cheaper: $1.43 against $67.50 per million tokens, blended at three input tokens per output token.
- Which has the larger context window?
- GPT-5.5 Pro accepts more context: 1.1M against 1M tokens.