# Mellum2-12B-A2.5B-Instruct benchmark scores

Other · model · https://benchgap.net/model/mellum2-12b-a2-5b-instruct · data retrieved 2026-11-22

As of 2026-11-22, Mellum2-12B-A2.5B-Instruct (Other) has measured scores on 6 benchmarks and estimated scores on 25 more.

| Benchmark | Score | Source |
|---|---|---|
| [AA-GPQA Diamond](https://benchgap.net/b/aa-gpqa-diamond/current.md) | 54.9% | estimated ± 4.4 pp, medium confidence |
| [AA-HLE](https://benchgap.net/b/aa-hle/current.md) | 0.0% | estimated ± 3.9 pp, medium confidence |
| [AA-SciCode](https://benchgap.net/b/aa-sci-code/current.md) | 9.1% | estimated ± 4.7 pp, medium confidence |
| [Artificial Analysis Intelligence Index](https://benchgap.net/b/artificialanalysis/current.md) | 9.4% | estimated ± 7.4 pp, low confidence |
| [CritPt](https://benchgap.net/b/critpt/current.md) | 0.3% | estimated ± 4.0 pp, high confidence |
| [GDPval-AA](https://benchgap.net/b/gdpvalaanormalized/current.md) | 0.0% | estimated ± 5.5 pp, low confidence |
| [GPQA Diamond](https://benchgap.net/b/gpqa-diamond/current.md) | 40.9% | measured |
| [HLE](https://benchgap.net/b/hle/current.md) | 0.6% | estimated ± 6.5 pp, low confidence |
| [IFBench](https://benchgap.net/b/ifbench/current.md) | 35.7% | estimated ± 9.2 pp, medium confidence |
| [LiveCodeBench v6](https://benchgap.net/b/livecodebench-v6/current.md) | 37.2% | measured |
| [AA-Omniscience Accuracy](https://benchgap.net/b/omniscienceaccuracy/current.md) | 2.3% | estimated ± 8.4 pp, low confidence |
| [SciCode](https://benchgap.net/b/scicode/current.md) | 11.0% | estimated ± 4.9 pp, medium confidence |
| [IFEval](https://benchgap.net/b/ifeval/current.md) | 75.8% | measured |
| [τ²-bench results](https://benchgap.net/b/tau2-bench/current.md) | 68.5% | estimated ± 12.4 pp, low confidence |
| [AA Coding Index](https://benchgap.net/b/aa-coding-index/current.md) | 3.1% | estimated ± 5.7 pp, low confidence |
| [SWE-bench Verified](https://benchgap.net/b/swe-bench-verified/current.md) | 57.5% | estimated ± 9.7 pp, low confidence |
| [BFCL v4](https://benchgap.net/b/bfcl-v4/current.md) | 44.2% | measured |
| [MMLU-Pro](https://benchgap.net/b/mmlu-pro/current.md) | 62.4% | estimated ± 1.1 pp, high confidence |
| [AA-IFBench](https://benchgap.net/b/aa-if-bench/current.md) | 20.2% | estimated ± 8.8 pp, medium confidence |
| [BBH](https://benchgap.net/b/bbh/current.md) | 82.7% | estimated ± 4.5 pp, medium confidence |
| [GPQA](https://benchgap.net/b/gpqa/current.md) | 40.9% | measured |
| [MMLU](https://benchgap.net/b/mmlu/current.md) | 82.7% | estimated ± 4.2 pp, high confidence |
| [Vals GPQA Diamond](https://benchgap.net/b/vals-gpqa-diamond/current.md) | 39.1% | estimated ± 4.5 pp, medium confidence |
| [Vals MMLU-Pro](https://benchgap.net/b/vals-mmlu-pro/current.md) | 61.6% | estimated ± 2.3 pp, medium confidence |
| [MMMLU](https://benchgap.net/b/mmmlu/current.md) | 75.4% | estimated ± 1.4 pp, low confidence |
| [C-Eval](https://benchgap.net/b/ceval/current.md) | 77.1% | estimated ± 1.2 pp, medium confidence |
| [MMLU-Redux](https://benchgap.net/b/mmluredux/current.md) | 78.1% | measured |
| [SuperGPQA](https://benchgap.net/b/supergpqa/current.md) | 36.0% | estimated ± 4.1 pp, high confidence |
| [MMLU-Pro (Arcee)](https://benchgap.net/b/mmlu-pro-arcee/current.md) | 61.2% | estimated ± 3.4 pp, low confidence |
| [SimpleQA](https://benchgap.net/b/simpleqa/current.md) | 14.6% | estimated ± 13.4 pp, low confidence |
| [PinchBench](https://benchgap.net/b/pinchbench/current.md) | 50.6% | estimated ± 5.0 pp, low confidence |
