# MAI-Thinking-1 vs Step 5 Preview

> Side-by-side benchmark comparison for MAI-Thinking-1 and Step 5 Preview across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/mai-thinking-1-vs-step-5-preview
- Last updated: September 21, 2026

- Shared sourced benchmarks: 1
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Evidence is too thin for a benchmark verdict: MAI-Thinking-1 and Step 5 Preview share only 1 sourced benchmark row.

## Summary

- Only 1 shared sourced benchmark row is available. That is not enough for a broad capability verdict, so this mirror does not name an overall winner.
- Step 5 Preview has the larger published context window at 1M.

## Model Snapshot

| Property | MAI-Thinking-1 | Step 5 Preview |
|----------|----------|----------|
| Creator | Microsoft | StepFun |
| Type | Proprietary | Pending |
| Reasoning | Reasoning | Reasoning |
| Context | 256K | 1M |
| Independent public score (not head-to-head) | 50.63 | null |
| Benchmarks Covered | 14 | 22 |
| Pricing (input/output) | N/A | $1.00 / $2.70 |

## Category Breakdown

### Agentic

- Winner: Insufficient shared evidence
- MAI-Thinking-1 public-lane score: 49.8 (Estimated · #60/154)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | MAI-Thinking-1 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 46% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 85.0% | Coming soon |
| Terminal-Bench 4.0 | Coming soon | 33.30% | Coming soon |
| CyberGym | Coming soon | 84.7% | 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 |
| JobBench | Coming soon | 59.0% | Coming soon |
| APEX-Agents | Coming soon | 37.8% | Coming soon |
| DRACO | Coming soon | 83.3% | Coming soon |
| BrowseComp | Coming soon | 88.7% | Coming soon |
| Agents' Last Exam | Coming soon | 29.5% | Coming soon |

### Coding

- Winner: Insufficient shared evidence
- MAI-Thinking-1 public-lane score: 50 (Estimated · #67/156)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | MAI-Thinking-1 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| LiveCodeBench v6 | 87.7% | Coming soon | Coming soon |
| SWE-bench Verified | 73.5% | Coming soon | Coming soon |
| SWE-bench Pro | 52.8% | Coming soon | Coming soon |
| Terminal-Bench 2.0 | 46.0% | 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: Insufficient shared evidence
- MAI-Thinking-1 public-lane score: Coming soon
- Step 5 Preview public-lane score: 78.5 (Unranked · 2 rankable rows)

| Benchmark | MAI-Thinking-1 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| Graphwalks BFS 128K | 90% | Coming soon | Coming soon |
| CritPt | Coming soon | 20.9% | Coming soon |

### Multimodal & Grounded

- Winner: Insufficient shared evidence
- MAI-Thinking-1 public-lane score: Coming soon
- Step 5 Preview public-lane score: 61.2 (#28/49)

| Benchmark | MAI-Thinking-1 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| OfficeQA Pro | Coming soon | 60.3% | Coming soon |
| MMMU-Pro | Coming soon | 76% | Coming soon |

### Knowledge

- Winner: Insufficient shared evidence
- MAI-Thinking-1 public-lane score: 51.5 (Estimated · #73/186)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | MAI-Thinking-1 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 84.2% | Coming soon | Coming soon |
| GPQA-D | 84.2% | 93.5% | Step 5 Preview |
| MMLU-Pro | 85% | Coming soon | Coming soon |
| SimpleQA | 31% | Coming soon | Coming soon |
| HLE | Coming soon | 46.5% | Coming soon |

### Instruction Following

- Winner: Insufficient shared evidence
- MAI-Thinking-1 public-lane score: 95.4 (#1/124)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | MAI-Thinking-1 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| IFBench | 85% | Coming soon | Coming soon |

### Mathematics

- Winner: Insufficient shared evidence
- MAI-Thinking-1 public-lane score: 72.8 (Unranked · 3 rankable rows)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | MAI-Thinking-1 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| AIME 2025 | 97% | Coming soon | Coming soon |
| AIME26 | 94.5% | Coming soon | Coming soon |
| HMMT Feb 2026 | 84.9% | Coming soon | Coming soon |

## FAQ

### Is there enough evidence to pick between MAI-Thinking-1 and Step 5 Preview?

Not yet. The models share only 1 sourced benchmark row, so this mirror reports the available measurements and metadata without naming a broad winner.

### What can I compare safely on this page?

Published pricing, context windows, model type, and individual shared benchmark rows are directly comparable. Independent overall scores and non-overlapping benchmark coverage are labeled separately and should not be read as a controlled head-to-head result.

## Related Comparisons

- [MAI-Thinking-1 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-mai-thinking-1)
- [Step 5 Preview vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-step-5-preview)
- [MAI-Thinking-1 vs GPT-6 Astra](/compare/gpt-6-astra-vs-mai-thinking-1)
- [Step 5 Preview vs GPT-6 Astra](/compare/gpt-6-astra-vs-step-5-preview)

## Explore More

- [MAI-Thinking-1 profile](/models/mai-thinking-1)
- [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)