# MiniMax M3 vs Step 5 Preview

> Side-by-side benchmark comparison for MiniMax M3 and Step 5 Preview across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/minimax-m3-vs-step-5-preview
- Last updated: September 21, 2026

- Shared sourced benchmarks: 4
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Pick Step 5 Preview if you want the stronger benchmark profile. MiniMax M3 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 61.13.
- The clearest category separation is in multimodal & grounded, where the averages are 52 for MiniMax M3 and 61.2 for Step 5 Preview.
- The biggest single benchmark swing is OfficeQA Pro in Multimodal & Grounded, with scores of 45.1% and 60.3%.
- MiniMax M3 is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.

## Model Snapshot

| Property | MiniMax M3 | Step 5 Preview |
|----------|----------|----------|
| Creator | MiniMax | StepFun |
| Type | Open Weight | Pending |
| Reasoning | Non-Reasoning | Reasoning |
| Context | 1M | 1M |
| Overall Score | 61.13 | null |
| Benchmarks Covered | 27 | 22 |
| Pricing (input/output) | $0.30 / $1.20 | $1.00 / $2.70 |

## Category Breakdown

### Agentic

- Winner: Coming soon
- MiniMax M3 public-lane score: 42.5 (Supported · #109/154)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | MiniMax M3 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 66% | Coming soon | Coming soon |
| BrowseComp | 83.5% | 88.7% | Step 5 Preview |
| OSWorld-Verified | 70.1% | Coming soon | Coming soon |
| MCP Atlas | 74.2% | 85.6% | Step 5 Preview |
| Claw-Eval | 74.5% | Coming soon | Coming soon |
| BankerToolBench | 76.1% | Coming soon | Coming soon |
| ResearchClawBench | 19.8% | Coming soon | Coming soon |
| OSWorld 2.0 | 4.6% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 53.6% | 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 |
| JobBench | Coming soon | 59.0% | 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
- MiniMax M3 public-lane score: 48.9 (Estimated · #71/156)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | MiniMax M3 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 80.5% | Coming soon | Coming soon |
| SWE-bench Pro | 59% | Coming soon | Coming soon |
| Terminal-Bench 2.0 | 66.0% | Coming soon | Coming soon |
| NL2Repo | 42.1% | Coming soon | Coming soon |
| VIBE V2 | 50.1% | Coming soon | Coming soon |
| SVG-Bench | 63.7% | Coming soon | Coming soon |
| KernelBench Hard | 28.8% | Coming soon | Coming soon |
| OpenHarmony Bench | 48.4% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 82.2% | Coming soon | Coming soon |
| SWE-bench (Vals) | 75.0% | 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: Coming soon
- MiniMax M3 public-lane score: 78 (#4/19)
- Step 5 Preview public-lane score: 78.5 (Unranked · 2 rankable rows)

| Benchmark | MiniMax M3 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| CritPt | Coming soon | 20.9% | Coming soon |

### Multimodal & Grounded

- Winner: Step 5 Preview
- MiniMax M3 public-lane score: 52 (#35/49)
- Step 5 Preview public-lane score: 61.2 (#28/49)

| Benchmark | MiniMax M3 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| OfficeQA Pro | 45.1% | 60.3% | Step 5 Preview |
| OmniDocBench 1.5 | 91.6% | Coming soon | Coming soon |
| MMMU-Pro | 78.1% | 76% | MiniMax M3 |
| VideoMMMU | 84.6% | Coming soon | Coming soon |
| Video-MME (with subtitle) | 85.4% | Coming soon | Coming soon |

### Knowledge

- Winner: Coming soon
- MiniMax M3 public-lane score: 53.2 (Supported · #65/186)
- Step 5 Preview public-lane score: Coming soon

| Benchmark | MiniMax M3 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| GPQA Diamond (Vals) | 92.7% | Coming soon | Coming soon |
| MMLU-Pro (Vals) | 84.2% | Coming soon | Coming soon |
| GPQA-D | Coming soon | 93.5% | Coming soon |
| HLE | Coming soon | 46.5% | Coming soon |

### Instruction Following

- Winner: Coming soon
- MiniMax M3 public-lane score: 92.4 (#5/124)
- Step 5 Preview public-lane score: Coming soon

### Mathematics

- Winner: Coming soon
- MiniMax M3 public-lane score: Coming soon
- Step 5 Preview public-lane score: Coming soon

| Benchmark | MiniMax M3 | Step 5 Preview | Winner |
|-----------|-----------|-----------|--------|
| USAMO 2026 | 85.7% | Coming soon | Coming soon |

## FAQ

### Which is better overall, MiniMax M3 or Step 5 Preview?

Step 5 Preview is ahead overall on BenchLM right now.

### Where is the biggest gap between MiniMax M3 and Step 5 Preview?

The widest category gap is in multimodal & grounded, where the averages are 52 for MiniMax M3 and 61.2 for Step 5 Preview.

### How many benchmarks does MiniMax M3 cover on BenchLM?

MiniMax M3 currently has 27 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

- [MiniMax M3 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-minimax-m3)
- [Step 5 Preview vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-step-5-preview)
- [MiniMax M3 vs GPT-6 Astra](/compare/gpt-6-astra-vs-minimax-m3)
- [Step 5 Preview vs GPT-6 Astra](/compare/gpt-6-astra-vs-step-5-preview)

## Explore More

- [MiniMax M3 profile](/models/minimax-m3)
- [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)