# MiMo-V2.5-Pro Benchmark Scores & Performance

> MiMo-V2.5-Pro by Xiaomi scores 65.41/100 overall, ranking #39 out of 411 AI models.

Canonical page: https://benchlm.ai/models/mimo-v2-5-pro

Last updated: September 4, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | Xiaomi |
| Source Type | Proprietary |
| Reasoning Type | Reasoning |
| Context Window | 1M |
| Overall Score | 65.41/100 |
| Overall Rank | #39 of 411 |

## Family & Coverage

- Family: MiMo-V2.5
- Variant: pro
- Benchmarks covered: 30 of 422
- Sibling models: [MiMo-V2.5](/models/mimo-v2-5)
- Related earlier model: [MiMo-V2-Pro](/models/mimo-v2-pro)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score |
|-----------|-------|
| [Claw-Eval](/benchmarks/claw-eval) | 63.8% |
| [GDPval-AA](/benchmarks/gdpvalaa) | 1265 |
| [τ³-bench results](/benchmarks/tau3-bench) | 72.9% |
| [Terminal-Bench 2.0](/benchmarks/terminal-bench-2) | 68.4% |
| [AA Agentic Index](/benchmarks/aaagenticindex) | 29.5% |
| [τ²-bench results](/benchmarks/tau2-bench) | 94.2% |
| [GDPval-AA](/benchmarks/gdpvalaanormalized) | 38.0% |
| [APEX-Agents-AA](/benchmarks/apexagentsaa) | 2.4% |
| [Gert Labs](/benchmarks/gertlabs) | 62.70% |
| [Terminal-Bench 2.1 (Vals)](/benchmarks/valsterminalbench21) | 57.3% |

## Coding Benchmarks

| Benchmark | Score |
|-----------|-------|
| [SWE-bench Pro](/benchmarks/swe-bench-pro) | 57.2% |
| [Terminal-Bench 2.0](/benchmarks/terminal-bench-2) | 68.4% |
| [AA Coding Index](/benchmarks/aacodingindex) | 60.2% |
| [AA-SciCode](/benchmarks/aascicode) | 50.2% |
| [LiveCodeBench (Vals)](/benchmarks/valslivecodebench) | 81.4% |
| [SWE-bench (Vals)](/benchmarks/valsswebench) | 74.0% |

## Multimodal & Grounded Benchmarks

| Benchmark | Score |
|-----------|-------|
| [Design Arena Website](/benchmarks/designarenawebsite) | 1289 |

## Reasoning Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA-LCR](/benchmarks/lcr) | 77.7% |
| [CritPt](/benchmarks/critpt) | 4.0% |

## Knowledge Benchmarks

| Benchmark | Score |
|-----------|-------|
| [HLE](/benchmarks/hle) | 48% |
| [HLE w/o tools](/benchmarks/hlenotools) | 34% |
| [Artificial Analysis Intelligence Index](/benchmarks/artificialanalysis) | 42.9% |
| [AA-GPQA Diamond](/benchmarks/aagpqadiamond) | 86.6% |
| [AA-HLE](/benchmarks/aahle) | 35.7% |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | 3.3% |
| [AA-Omniscience Accuracy](/benchmarks/omniscienceaccuracy) | 22.4% |
| [AA-Omniscience Hallucination Rate](/benchmarks/omnisciencehallucinationrate) | 24.7% |
| [GPQA Diamond (Vals)](/benchmarks/valsgpqadiamond) | 82.6% |
| [MMLU-Pro (Vals)](/benchmarks/valsmmlupro) | 84.6% |

## Instruction Following Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA-IFBench](/benchmarks/aaifbench) | 79.9% |

## Other Xiaomi Models

- [MiMo-V2-Pro](/models/mimo-v2-pro) - Score: 64.89
- [MiMo-V2.5](/models/mimo-v2-5) - Score: 60.72
- [MiMo-V2-Omni](/models/mimo-v2-omni) - Score: 60.62
- [MiMo-V2-Flash](/models/mimo-v2-flash) - Score: 52.18