# MiniMax M2.7 Benchmark Scores & Performance

> MiniMax M2.7 by MiniMax scores 55.14/100 overall, ranking #88 out of 486 AI models.

Canonical page: https://benchlm.ai/models/minimax-m2-7

Last updated: September 15, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | MiniMax |
| Source Type | Open Weight |
| Reasoning Type | Non-Reasoning |
| Context Window | 200K |
| Overall Score | 55.14/100 |
| Overall Rank | #88 of 486 |

## Family & Coverage

- Family: MiniMax M2.7
- Variant: base
- Benchmarks covered: 40 of 439
- Related earlier model: [MiniMax M2.5](/models/minimax-m2-5)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score |
|-----------|-------|
| [Terminal-Bench 2.0](/benchmarks/terminal-bench-2) | 57% |
| [τ²-bench results](/benchmarks/tau2-bench) | 84.8% |
| [Toolathlon](/benchmarks/toolathlon) | 46.3% |
| [MLE-Bench Lite](/benchmarks/mle-bench-lite) | 66.6% |
| [MM-ClawBench](/benchmarks/mmclawbench) | 62.7% |
| [Claw-Eval](/benchmarks/claw-eval) | 48.7% |
| [APEX-Agents-AA](/benchmarks/apexagentsaa) | 10.6% |
| [GDPval-AA](/benchmarks/gdpvalaanormalized) | 29.4% |
| [Gert Labs](/benchmarks/gertlabs) | 40.40% |
| [Terminal-Bench 2.1 (Vals)](/benchmarks/valsterminalbench21) | 48.7% |
| [AA Agentic Index](/benchmarks/aaagenticindex) | 16.8% |
| [GDPval-AA](/benchmarks/gdpvalaa) | 1087 |

## Coding Benchmarks

| Benchmark | Score |
|-----------|-------|
| [SWE-bench Verified*](/benchmarks/sweverifiedarcee) | 75.4% |
| [SWE-bench Pro](/benchmarks/swe-bench-pro) | 56.2% |
| [SWE-Rebench](/benchmarks/swe-rebench) | 51.9% |
| [SWE Multilingual](/benchmarks/swe-bench-multilingual) | 76.5% |
| [Multi-SWE Bench](/benchmarks/multiswebench) | 52.7% |
| [VIBE-Pro](/benchmarks/vibepro) | 55.6% |
| [NL2Repo](/benchmarks/nl2repo) | 39.8% |
| [Vibe Code Bench](/benchmarks/vibecodebench) | 27.04% |
| [React Native Evals](/benchmarks/reactnativeevals) | 71.4% |
| [AA-SciCode](/benchmarks/aascicode) | 50.1% |
| [LiveCodeBench (Vals)](/benchmarks/valslivecodebench) | 79.9% |
| [SWE-bench (Vals)](/benchmarks/valsswebench) | 73.8% |
| [AA Coding Index](/benchmarks/aacodingindex) | 52.6% |

## Multimodal & Grounded Benchmarks

| Benchmark | Score |
|-----------|-------|
| [Design Arena Website](/benchmarks/designarenawebsite) | 1257 |

## Reasoning Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA-LCR](/benchmarks/lcr) | 78.3% |
| [CritPt](/benchmarks/critpt) | 0.6% |

## Knowledge Benchmarks

| Benchmark | Score |
|-----------|-------|
| [GPQA-D](/benchmarks/gpqa-diamond) | 87.0% |
| [MMLU-Pro (Arcee)](/benchmarks/mmluproarcee) | 80.8% |
| [Artificial Analysis Intelligence Index](/benchmarks/artificialanalysis) | 23.2% |
| [AA-GPQA Diamond](/benchmarks/aagpqadiamond) | 87.4% |
| [AA-HLE](/benchmarks/aahle) | 29.6% |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | 0.8% |
| [AA-Omniscience Accuracy](/benchmarks/omniscienceaccuracy) | 26.8% |
| [AA-Omniscience Hallucination Rate](/benchmarks/omnisciencehallucinationrate) | 35.6% |
| [GPQA Diamond (Vals)](/benchmarks/valsgpqadiamond) | 86.6% |
| [MMLU-Pro (Vals)](/benchmarks/valsmmlupro) | 80.4% |

## Instruction Following Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA-IFBench](/benchmarks/aaifbench) | 75.7% |

## Mathematics Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AIME25 (Arcee)](/benchmarks/aime2025arcee) | 80.0% |

## Other MiniMax Models

- [MiniMax M3](/models/minimax-m3) - Score: 61.47
- [MiniMax M2.5](/models/minimax-m2-5) - Score: 54.89
- [MiniMax M1 80k](/models/minimax-m1-80k) - Score: 21.56
- [MiniMax Speech-02 HD](/models/minimax-speech-02-hd) - Score: not computed