Claude Sonnet 5.5 vs Step 5 Preview
Claude Sonnet 5.5 
Step 5 Preview Verdict
- Claude Sonnet 5.5 (max) leads on quality: 67.6 vs 62.4.
- Claude Sonnet 5.5 (max) is stronger in agents & tools, coding, maths, reasoning.
- Step 5 Preview is stronger in composite, knowledge, long context, human preference, multimodal.
- Step 5 Preview is 2.8× cheaper ($1.43 vs $4.00 per 1M blended).
- Claude Sonnet 5.5 (max) streams 1.6× faster (141 vs 87 tokens per second).
| Metric | Claude Sonnet 5.5 (max) | Step 5 Preview |
|---|---|---|
| BenchLeader Index | 67.6 | 62.4 |
| Agents & tools score | 69.5 | 47.2 |
| Coding score | 69.8 | 65.6 |
| Composite score | 72.6 | 79.1 |
| Knowledge score | 65.4 | 69.1 |
| Long context score | 66.6 | 69.5 |
| Maths score | 72.9 | 61.1 |
| Reasoning score | 81.1 | 72.2 |
| Human preference score | – | 63.6 |
| Multimodal score | – | 60.0 |
| Blended price $/M | $4.00 | $1.43 |
| Output speed | 141 tok/s | 87 tok/s |
| Time to first answer | 487.1 s | 25.9 s |
| Context window | 1M | 1M |
| GPQA Diamond | 95.6% | – |
| FrontierMath Tiers 1–3 | 88.8% | – |
| FrontierMath Tier 4 | 80.5% | – |
| OTIS Mock AIME | 100.0% | – |
| SimpleQA Verified | 46.5% | – |
| Terminal-Bench | 61.8% | – |
| SciCode | 61.0% | 58.9% |
| APEX-Agents | 75.5% | – |
| ProofBench | 100.0% | 42.0% |
| LMArena Text | – | 1447 |
| LMArena Hard Prompts | – | 1479 |
| LMArena Coding | – | 1503 |
| LMArena Vision | – | 1267 |
| LMArena Agent | 12 | 0 |
| LiveBench | 75.7% | – |
| LiveBench Reasoning | 91.6% | – |
| LiveBench Coding | 91.4% | – |
| LiveBench Agentic Coding | 56.3% | – |
| LiveBench Mathematics | 96.1% | – |
| LiveBench Data Analysis | 59.5% | – |
| LiveBench Language | 78.0% | – |
| LiveBench Instruction Following | 56.8% | – |
| AA Intelligence Index v4.3.2 | 56 | 43.7 |
| AA-LCR | 82.7% | 88.3% |
| MMMU-Pro | – | 76.4% |
| AA-Omniscience | 32.3 | 16.4 |
| Humanity's Last Exam (AA) | 55.0% | 46.5% |
| SciCode (AA) | 61.0% | 58.9% |
| CritPt | 31.4% | 20.9% |
| GDPval-AA v2.1 | 67.0% | 54.2% |
| ITBench SRE (AA) | – | 55.6% |
| Analyst Agent (AA) | 57.5% | 35.0% |
| APEX-Agents (AA) | – | 38.0% |
| LMArena Maths | – | 1478 |
| LMArena Creative Writing | – | 1406 |
| LMArena Instruction Following | – | 1449 |
| LMArena Multi-turn | – | 1453 |
| LMArena Longer Queries | – | 1462 |
| Mystery Game Puzzles | 65.0% | – |
| CursorBench | 55.5% | – |
| FrontierSWE | 61.9% | – |
| Terminal-Bench 4.0 (AA) | 63.6% | 33.3% |
| AutomationBench | 71.8% | 51.0% |
| GDP.pdf | 25.8% | 14.8% |
| MLCR | 75.0% | 16.7% |
| Harvey LAB | 2.8% | – |
| EnterpriseOps-Gym | – | 47.2% |
| AA-Omniscience: accuracy | 54.0% | 41.5% |
| AA-Omniscience: non-hallucination | 53.0% | 57.0% |
| AA-Briefcase v1.1 | 1823 | 1425 |
Data as of 2026-10-11. Best configuration of each model; every score links to its source on the model pages.
Claude Sonnet 5.5 vs Step 5 Preview: questions
- Is Claude Sonnet 5.5 better than Step 5 Preview?
- Claude Sonnet 5.5 (max) leads on quality: 67.6 vs 62.4. The BenchLeader Index combines every independent quality benchmark; Claude Sonnet 5.5 (max) is ahead overall as of 2026-10-11, but check the category scores for your use.
- Is Claude Sonnet 5.5 better than Step 5 Preview for coding?
- Claude Sonnet 5.5 scores higher in coding (70 vs 66 on the category index, where 50 is average).
- Is Claude Sonnet 5.5 better than Step 5 Preview for agentic tasks?
- Claude Sonnet 5.5 scores higher in agentic tasks (70 vs 47 on the category index, where 50 is average).
- Which is cheaper, Claude Sonnet 5.5 or Step 5 Preview?
- Step 5 Preview is cheaper: $1.43 against $4.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Sonnet 5.5 or Step 5 Preview?
- Claude Sonnet 5.5 streams faster: 141 against 87 output tokens per second.
- Which has the larger context window?
- Both accept 1M tokens of context.