# Claude Opus 4.6 vs Step 5 Preview

> Side-by-side benchmark comparison for Claude Opus 4.6 and Step 5 Preview across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/claude-opus-4-6-vs-step-5-preview
- Last updated: September 21, 2026

- Shared sourced benchmarks: 6
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Pick Step 5 Preview if you want the stronger benchmark profile. Claude Opus 4.6 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- Step 5 Preview leads overall 72 to 69.42.
- The clearest category separation is in reasoning, where the averages are 67.1 for Claude Opus 4.6 and 78.5 for Step 5 Preview.
- The biggest single benchmark swing is JobBench in Agentic, with scores of 36.7% and 59.0%.
- Step 5 Preview is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.

## Model Snapshot

| Property | Claude Opus 4.6 | Step 5 Preview |
|----------|----------|----------|
| Creator | Anthropic | StepFun |
| Type | Proprietary | Pending |
| Reasoning | Non-Reasoning | Reasoning |
| Context | 1M | 1M |
| Overall Score | 69.42 | null |
| Benchmarks Covered | 34 | 22 |
| Pricing (input/output) | $5.00 / $25.00 | $1.00 / $2.70 |

## Category Breakdown

### Agentic

- Winner: Coming soon
- Claude Opus 4.6 public-lane score: 50.1 (Supported · #58/154)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | Claude Opus 4.6 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 65.4% | Coming soon | Coming soon |
| BrowseComp | 83.7% | 88.7% | Step 5 Preview |
| OSWorld-Verified | 72.7% | Coming soon | Coming soon |
| Claw-Eval | 70.4% | Coming soon | Coming soon |
| DeepSearchQA | 73.7% | Coming soon | Coming soon |
| CyberGym | 66.6% | 84.7% | Step 5 Preview |
| Gert Labs | 61.85% | Coming soon | Coming soon |
| ResearchClawBench | 19.9% | Coming soon | Coming soon |
| JobBench | 36.7% | 59.0% | Step 5 Preview |
| ApprenticeBench | 5% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 85.0% | Coming soon |
| Terminal-Bench 4.0 | Coming soon | 33.30% | Coming soon |
| AutomationBench | Coming soon | 44.0% | Coming soon |
| Toolathlon-Verified | Coming soon | 74.1% | Coming soon |
| MCP Atlas | Coming soon | 85.6% | Coming soon |
| APEX-Agents | Coming soon | 37.8% | Coming soon |
| DRACO | Coming soon | 83.3% | Coming soon |
| Agents' Last Exam | Coming soon | 29.5% | Coming soon |

### Coding

- Winner: Coming soon
- Claude Opus 4.6 public-lane score: 56.8 (Supported · #35/156)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | Claude Opus 4.6 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 80.8% | Coming soon | Coming soon |
| SWE-bench Verified* | 75.6% | Coming soon | Coming soon |
| LiveCodeBench Pro | 70.7% | Coming soon | Coming soon |
| SWE-bench Pro | 53.4% | Coming soon | Coming soon |
| SWE-Rebench | 65.3% | Coming soon | Coming soon |
| React Native Evals | 84.1% | Coming soon | Coming soon |
| Vibe Code Bench | 57.57% | Coming soon | Coming soon |
| FrontierCode 1.1 Main | 26.9% | Coming soon | Coming soon |
| DeepSWE | Coming soon | 67.7% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 85.0% | Coming soon |
| SciCode | Coming soon | 58.9% | Coming soon |
| ProgramBench | Coming soon | 80.5% | Coming soon |
| sweMarathon | Coming soon | 72.7% | Coming soon |
| MLS-Bench Lite | Coming soon | 40.5% | Coming soon |

### Reasoning

- Winner: Coming soon
- Claude Opus 4.6 public-lane score: 67.1 (Unranked · 2 rankable rows)
- Step 5 Preview public-lane score: 78.5 (Unranked · 2 rankable rows)

| Benchmark | Claude Opus 4.6 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| CritPt | Coming soon | 20.9% | Coming soon |

### Multimodal & Grounded

- Winner: Coming soon
- Claude Opus 4.6 public-lane score: 59.6 (#29/49)
- Step 5 Preview public-lane score: 61.2 (#28/49)

| Benchmark | Claude Opus 4.6 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 77.3% | 76% | Claude Opus 4.6 |
| ERQA | 51.6% | Coming soon | Coming soon |
| ScreenSpot Pro | 83.1% | Coming soon | Coming soon |
| MedXpertQA (MM) | 64.8% | Coming soon | Coming soon |
| OfficeQA Pro | Coming soon | 60.3% | Coming soon |

### Knowledge

- Winner: Coming soon
- Claude Opus 4.6 public-lane score: 61.2 (Estimated · #34/186)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | Claude Opus 4.6 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 91.3% | Coming soon | Coming soon |
| GPQA-D | 89.2% | 93.5% | Step 5 Preview |
| SuperGPQA | 95% | Coming soon | Coming soon |
| MMLU-Pro | 82% | Coming soon | Coming soon |
| MMLU-Pro (Arcee) | 89.1% | Coming soon | Coming soon |
| HLE | 53% | 46.5% | Claude Opus 4.6 |
| HLE w/o tools | 40% | Coming soon | Coming soon |
| HealthBench Hard | 14.8% | Coming soon | Coming soon |
| MedXpertQA (Text) | 52.1% | Coming soon | Coming soon |

### Instruction Following

- Winner: Coming soon
- Claude Opus 4.6 public-lane score: 51 (#79/124)
- Step 5 Preview public-lane score: Coming soon

### Mathematics

- Winner: Coming soon
- Claude Opus 4.6 public-lane score: 58.5 (Unranked · 3 rankable rows)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | Claude Opus 4.6 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| AIME25 (Arcee) | 99.8% | Coming soon | Coming soon |
| FrontierMath v2 (Tiers 1-3) | 40.700% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | 22.900% | Coming soon | Coming soon |

## FAQ

### Which is better overall, Claude Opus 4.6 or Step 5 Preview?

Step 5 Preview is ahead overall on BenchLM right now.

### Where is the biggest gap between Claude Opus 4.6 and Step 5 Preview?

The widest category gap is in reasoning, where the averages are 67.1 for Claude Opus 4.6 and 78.5 for Step 5 Preview.

### How many benchmarks does Claude Opus 4.6 cover on BenchLM?

Claude Opus 4.6 currently has 34 sourced benchmark scores on BenchLM.

### How many benchmarks does Step 5 Preview cover on BenchLM?

Step 5 Preview currently has 22 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Claude Opus 4.6 vs Claude Opus 4.6 (Adaptive)](/compare/claude-opus-4-6-vs-claude-opus-4-6-thinking)
- [Step 5 Preview vs Claude Opus 4.6 (Adaptive)](/compare/claude-opus-4-6-thinking-vs-step-5-preview)
- [Claude Opus 4.6 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-claude-opus-4-6)
- [Step 5 Preview vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-step-5-preview)

## Explore More

- [Claude Opus 4.6 profile](/models/claude-opus-4-6)
- [Step 5 Preview profile](/models/step-5-preview)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)