Other · model
Mellum2-12B-A2.5B-Instruct benchmark scores
As of 2026-10-07, Mellum2-12B-A2.5B-Instruct (Other) has measured scores on 6 benchmarks and estimated scores on 32 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 39.1% | estimated ± 5.4 pp, low confidence |
| AA-HLE | 0.0% | estimated ± 4.0 pp, medium confidence |
| AA-SciCode | 1.1% | estimated ± 3.7 pp, medium confidence |
| Artificial Analysis Intelligence Index | 8.9% | estimated ± 4.2 pp, low confidence |
| CritPt | 0.6% | estimated ± 4.1 pp, medium confidence |
| GDPval-AA | 0.0% | estimated ± 5.8 pp, low confidence |
| GPQA Diamond | 40.9% | measured |
| HLE | 0.0% | estimated ± 5.9 pp, low confidence |
| IFBench | 34.4% | estimated ± 9.7 pp, medium confidence |
| LiveCodeBench v6 | 37.2% | measured |
| AA-Omniscience Accuracy | 5.1% | estimated ± 4.5 pp, low confidence |
| SciCode | 11.4% | estimated ± 5.4 pp, low confidence |
| IFEval | 75.8% | measured |
| τ²-bench results | 68.2% | estimated ± 14.1 pp, low confidence |
| AA Coding Index | 0.6% | estimated ± 6.4 pp, low confidence |
| SWE-bench Verified | 38.5% | estimated ± 10.5 pp, low confidence |
| BFCL v4 | 44.2% | measured |
| SWE-bench Pro | 0.1% | estimated ± 4.7 pp, medium confidence |
| MMLU-Pro | 60.4% | estimated ± 1.4 pp, high confidence |
| GPQA | 40.9% | measured |
| AA-IFBench | 15.0% | estimated ± 8.2 pp, low confidence |
| ARC-AGI-1 | 3.0% | estimated ± 2.7 pp, medium confidence |
| ARC-AGI-2 | 6.1% | estimated ± 10.1 pp, low confidence |
| Vals GPQA Diamond | 38.8% | estimated ± 2.0 pp, medium confidence |
| Vals MMLU-Pro | 0.0% | estimated ± 1.2 pp, medium confidence |
| HLE w/o tools | 5.7% | estimated ± 4.4 pp, medium confidence |
| C-Eval | 56.8% | estimated ± 0.9 pp, low confidence |
| Claw-Eval | 53.9% | estimated ± 2.3 pp, low confidence |
| MMLU-Redux | 78.1% | measured |
| NL2Repo | 22.2% | estimated ± 3.1 pp, low confidence |
| SuperGPQA | 35.9% | estimated ± 4.5 pp, high confidence |
| HealthBench Hard | 0.0% | estimated ± 5.1 pp, low confidence |
| MedXpertQA (Text) | 7.8% | estimated ± 4.5 pp, low confidence |
| MMLU-Pro (Arcee) | 61.2% | estimated ± 3.4 pp, low confidence |
| ARC-AGI-3 | 0.0% | estimated ± 2.1 pp, medium confidence |
| HealthBench Professional | 0.8% | estimated ± 4.8 pp, low confidence |
| MMMLU | 11.7% | estimated ± 1.1 pp, low confidence |
| MMLU | 62.5% | estimated ± 3.6 pp, low confidence |