# MiniCPM5-2B Benchmark Scores & Performance

> MiniCPM5-2B by OpenBMB has 29 source-displayable benchmark rows but no public overall score. It remains unranked.

Canonical page: https://benchlm.ai/models/minicpm5-2b

Last updated: September 10, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | OpenBMB |
| Source Type | Open Weight |
| Reasoning Type | Reasoning |
| Context Window | 131K |
| Official model card | [OpenBMB MiniCPM5-2B model card](https://huggingface.co/openbmb/MiniCPM5-2B) |
| Overall Score | Coming soon |
| Overall Rank | Unranked |

## Family & Coverage

- Family: MiniCPM5
- Variant: 2b
- Benchmarks covered: 29 of 434
- Sibling models: [MiniCPM5-1B](/models/minicpm5-1b)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score |
|-----------|-------|
| [BFCL v4](/benchmarks/bfcl-v4) | 66.6% |
| [Terminal-Bench 2.1](/benchmarks/terminalbench21) | 8.6% |
| [GDPval-AA](/benchmarks/gdpvalaanormalized) | 16.4% |

## Coding Benchmarks

| Benchmark | Score |
|-----------|-------|
| [LiveCodeBench v6](/benchmarks/livecodebench-v6) | 69.1% |
| [SciCode](/benchmarks/scicode) | 26.3% |
| [SWE-bench Verified](/benchmarks/swe-bench-verified) | 46.4% |
| [SWE-bench Pro](/benchmarks/swe-bench-pro) | 14.4% |
| [Terminal-Bench 2.1](/benchmarks/terminalbench21) | 8.6% |
| [AA-SciCode](/benchmarks/aascicode) | 26.3% |

## Reasoning Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA-LCR](/benchmarks/lcr) | 59.0% |
| [LongBench v2](/benchmarks/longbench-v2) | 43.7% |
| [CritPt](/benchmarks/critpt) | 0.3% |

## Knowledge Benchmarks

| Benchmark | Score |
|-----------|-------|
| [MMLU-Pro](/benchmarks/mmlu-pro) | 70.8% |
| [MMLU-Redux](/benchmarks/mmluredux) | 84.7% |
| [HLE](/benchmarks/hle) | 8.9% |
| [GPQA-D](/benchmarks/gpqa-diamond) | 70.2% |
| [SuperGPQA](/benchmarks/supergpqa) | 40.8% |
| [Artificial Analysis Intelligence Index](/benchmarks/artificialanalysis) | 13.1% |
| [AA-GPQA Diamond](/benchmarks/aagpqadiamond) | 70.2% |
| [AA-HLE](/benchmarks/aahle) | 8.9% |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | -11.6% |
| [AA-Omniscience Accuracy](/benchmarks/omniscienceaccuracy) | 8.4% |
| [AA-Omniscience Hallucination Rate](/benchmarks/omnisciencehallucinationrate) | 21.9% |

## Instruction Following Benchmarks

| Benchmark | Score |
|-----------|-------|
| [IFBench](/benchmarks/ifbench) | 66.3% |
| [IFEval](/benchmarks/ifeval) | 86.7% |

## Mathematics Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AIME 2025](/benchmarks/aime2025) | 86.5% |
| [AIME26](/benchmarks/aime2026) | 86.5% |
| [HMMT Feb 2026](/benchmarks/hmmtfeb2026) | 63.8% |
| [MATH-500](/benchmarks/math-500) | 94.6% |

## Other OpenBMB Models

- [MiniCPM5-1B](/models/minicpm5-1b) - Score: not computed
- [MiniCPM-o 2.6](/models/minicpm-o-2-6) - Score: not computed