# Qwen3.8 Max vs Step 5 Preview

> Side-by-side benchmark comparison for Qwen3.8 Max and Step 5 Preview across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/qwen3-8-max-vs-step-5-preview
- Last updated: September 21, 2026

- Shared sourced benchmarks: 11
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Pick Qwen3.8 Max if you want the stronger benchmark profile. Step 5 Preview only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- Qwen3.8 Max leads overall 73.19 to 72.
- The clearest category separation is in multimodal & grounded, where the averages are 87.4 for Qwen3.8 Max and 61.2 for Step 5 Preview.
- The biggest single benchmark swing is Agents' Last Exam in Agentic, with scores of 52.4% and 29.5%.
- Qwen3.8 Max is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.

## Model Snapshot

| Property | Qwen3.8 Max | Step 5 Preview |
|----------|----------|----------|
| Creator | Alibaba | StepFun |
| Type | Open Weight | Pending |
| Reasoning | Reasoning | Reasoning |
| Context | 1M | 1M |
| Overall Score | 73.19 | null |
| Benchmarks Covered | 60 | 22 |
| Pricing (input/output) | Pricing unavailable / Pricing unavailable | $1.00 / $2.70 |

## Category Breakdown

### Agentic

- Winner: Coming soon
- Qwen3.8 Max public-lane score: 67.5 (Supported · #9/154)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | Qwen3.8 Max | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.1 | 86.6% | 85.0% | Qwen3.8 Max |
| CoWorkBench | 74.8% | Coming soon | Coming soon |
| JobBench | 53.4% | 59.0% | Step 5 Preview |
| skillsBench | 70.2% | Coming soon | Coming soon |
| Agents' Last Exam | 52.4% | 29.5% | Qwen3.8 Max |
| AutomationBench | 27.3% | 44.0% | Step 5 Preview |
| Toolathlon-Verified | 72.5% | 74.1% | Step 5 Preview |
| WideResearch | 81.9% | Coming soon | Coming soon |
| HLE w/ tools | 56.2% | Coming soon | Coming soon |
| OSWorld-Verified | 86.1% | Coming soon | Coming soon |
| OSWorld 2.0 | 19.4% | Coming soon | Coming soon |
| WebArena-Verified | 66.8% | Coming soon | Coming soon |
| AndroidWorld | 85.3% | Coming soon | Coming soon |
| MobileWorld | 77.8% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 67.4% | Coming soon | Coming soon |
| Terminal-Bench 4.0 | Coming soon | 33.30% | Coming soon |
| CyberGym | Coming soon | 84.7% | 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 |
| BrowseComp | Coming soon | 88.7% | Coming soon |

### Coding

- Winner: Coming soon
- Qwen3.8 Max public-lane score: 60.9 (Supported · #21/156)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | Qwen3.8 Max | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.1 | 86.6% | 85.0% | Qwen3.8 Max |
| SWE-bench Pro | 67.7% | Coming soon | Coming soon |
| DeepSWE | 56.6% | 67.7% | Step 5 Preview |
| NL2Repo | 55.9% | Coming soon | Coming soon |
| FrontierSWE | 73.5% | Coming soon | Coming soon |
| MLS-Bench Lite | 41.0% | 40.5% | Qwen3.8 Max |
| PaperBench | 93.0% | Coming soon | Coming soon |
| VulcanBench v3 | 81.2% | Coming soon | Coming soon |
| OpenHarmony Bench | 60.8% | Coming soon | Coming soon |
| FrontierSWE v2 | 15.8% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 87.9% | Coming soon | Coming soon |
| SWE-bench (Vals) | 85.6% | Coming soon | Coming soon |
| SciCode | Coming soon | 58.9% | Coming soon |
| ProgramBench | Coming soon | 80.5% | Coming soon |
| sweMarathon | Coming soon | 72.7% | Coming soon |

### Reasoning

- Winner: Coming soon
- Qwen3.8 Max public-lane score: 86.6 (Unranked · 2 rankable rows)
- Step 5 Preview public-lane score: 78.5 (Unranked · 2 rankable rows)

| Benchmark | Qwen3.8 Max | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| MRCRv2 | 92.9% | Coming soon | Coming soon |
| LongBench v2 | 66.3% | Coming soon | Coming soon |
| CritPt | Coming soon | 20.9% | Coming soon |

### Multimodal & Grounded

- Winner: Coming soon
- Qwen3.8 Max public-lane score: 87.4 (#5/49)
- Step 5 Preview public-lane score: 61.2 (#28/49)

| Benchmark | Qwen3.8 Max | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 82.3% | 76% | Qwen3.8 Max |
| MathVision | 95.2% | Coming soon | Coming soon |
| MathVision w/ Python | 97.7% | Coming soon | Coming soon |
| BabyVision | 82.0% | Coming soon | Coming soon |
| BabyVision w/ Python | 91.3% | Coming soon | Coming soon |
| ZeroBench | 24.0% | Coming soon | Coming soon |
| ZeroBench w/ Python | 49.0% | Coming soon | Coming soon |
| MedXpertQA (MM) | 80.4% | Coming soon | Coming soon |
| ScreenSpot Pro | 84.5% | Coming soon | Coming soon |
| Vision2Web | 69.0% | Coming soon | Coming soon |
| CharXiv w/o tools | 88.4% | Coming soon | Coming soon |
| CharXiv | 93.5% | Coming soon | Coming soon |
| OmniDocBench 1.5 | 92.1% | Coming soon | Coming soon |
| OCRBench V2 | 74.2% | Coming soon | Coming soon |
| CC-OCR | 79.6% | Coming soon | Coming soon |
| RealWorldQA | 88.0% | Coming soon | Coming soon |
| ERQA | 77.8% | Coming soon | Coming soon |
| SimpleVQA | 75.0% | Coming soon | Coming soon |
| PerceptionBench | 63.5% | Coming soon | Coming soon |
| Video-MME (with subtitle) | 90.4% | Coming soon | Coming soon |
| VideoMMMU | 88.7% | Coming soon | Coming soon |
| MMVU | 82.4% | Coming soon | Coming soon |
| MLVU (M-Avg) | 90.8% | Coming soon | Coming soon |
| LVBench | 81.8% | Coming soon | Coming soon |
| OfficeQA Pro | Coming soon | 60.3% | Coming soon |

### Knowledge

- Winner: Coming soon
- Qwen3.8 Max public-lane score: 68.9 (Supported · #17/186)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | Qwen3.8 Max | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 92.6% | Coming soon | Coming soon |
| GPQA-D | 92.6% | 93.5% | Step 5 Preview |
| HLE | 43.6% | 46.5% | Step 5 Preview |
| HLE w/o tools | 43.6% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 93.7% | Coming soon | Coming soon |
| MMLU-Pro (Vals) | 88.6% | Coming soon | Coming soon |

### Instruction Following

- Winner: Coming soon
- Qwen3.8 Max public-lane score: 90.5 (#16/124)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | Qwen3.8 Max | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| IFBench | 82.8% | Coming soon | Coming soon |

## FAQ

### Which is better overall, Qwen3.8 Max or Step 5 Preview?

Qwen3.8 Max is ahead overall on BenchLM right now.

### Where is the biggest gap between Qwen3.8 Max and Step 5 Preview?

The widest category gap is in multimodal & grounded, where the averages are 87.4 for Qwen3.8 Max and 61.2 for Step 5 Preview.

### How many benchmarks does Qwen3.8 Max cover on BenchLM?

Qwen3.8 Max currently has 60 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

- [Qwen3.8 Max vs Qwen3.8 Max Preview](/compare/qwen3-8-max-vs-qwen3-8-max-preview)
- [Step 5 Preview vs Qwen3.8 Max Preview](/compare/qwen3-8-max-preview-vs-step-5-preview)
- [Qwen3.8 Max vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-qwen3-8-max)
- [Step 5 Preview vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-step-5-preview)

## Explore More

- [Qwen3.8 Max profile](/models/qwen3-8-max)
- [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)