Other · model
Ternary Bonsai 2 27B benchmark scores
As of 2026-11-22, Ternary Bonsai 2 27B (Other) has measured scores on 15 benchmarks and estimated scores on 72 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 84.7% | estimated ± 4.4 pp, high confidence |
| AA-HLE | 26.2% | estimated ± 3.9 pp, high confidence |
| AA-SciCode | 39.1% | estimated ± 3.8 pp, high confidence |
| AIME26 | 95.8% | measured |
| Artificial Analysis Intelligence Index | 17.3% | estimated ± 6.3 pp, low confidence |
| CritPt | 2.1% | estimated ± 4.0 pp, high confidence |
| GDPval-AA | 13.4% | estimated ± 11.0 pp, low confidence |
| GPQA Diamond | 85.8% | measured |
| HLE | 25.7% | estimated ± 6.5 pp, medium confidence |
| IFBench | 74.0% | measured |
| AA-LCR | 77.0% | measured |
| LiveCodeBench v6 | 90.1% | measured |
| AA-Omniscience Accuracy | 24.3% | estimated ± 8.4 pp, medium confidence |
| SciCode | 38.4% | estimated ± 3.6 pp, high confidence |
| Terminal-Bench 2.1 | 52.8% | measured |
| BrowseComp | 66.9% | estimated ± 13.0 pp, low confidence |
| HLE w/ tools | 32.3% | estimated ± 5.5 pp, low confidence |
| IFEval | 91.3% | measured |
| LongBench v2 | 63.2% | estimated ± 5.5 pp, medium confidence |
| τ²-bench results | 80.2% | measured |
| AA Coding Index | 28.8% | estimated ± 5.3 pp, medium confidence |
| AA-MMMU-Pro | 74.4% | estimated ± 2.4 pp, medium confidence |
| APEX-Agents-AA | 1.6% | estimated ± 11.4 pp, low confidence |
| SWE-bench Verified | 60.8% | measured |
| BFCL v4 | 61.5% | estimated ± 8.5 pp, low confidence |
| SWE-bench Pro | 37.2% | estimated ± 4.5 pp, high confidence |
| AA Agentic Index | 19.5% | estimated ± 12.7 pp, low confidence |
| MMLU-Pro | 79.0% | estimated ± 1.1 pp, high confidence |
| AA-IFBench | 62.1% | estimated ± 7.4 pp, medium confidence |
| AIME 2024 | 94.5% | estimated ± 2.6 pp, high confidence |
| GPQA | 85.8% | measured |
| MMLU | 88.1% | estimated ± 4.2 pp, medium confidence |
| Vals GPQA Diamond | 85.1% | estimated ± 2.1 pp, high confidence |
| Vals LiveCodeBench | 86.1% | estimated ± 3.4 pp, medium confidence |
| Vals MMLU-Pro | 84.3% | estimated ± 1.6 pp, high confidence |
| MATH-500 | 98.8% | measured |
| MMMLU | 88.1% | estimated ± 1.1 pp, medium confidence |
| MMMU | 75.8% | estimated ± 9.0 pp, medium confidence |
| AIME 2025 | 95.0% | measured |
| AA MMLU-Pro | 86.6% | estimated ± 1.2 pp, low confidence |
| ARC-AGI-1 | 82.2% | estimated ± 2.5 pp, medium confidence |
| ARC-AGI-2 | 33.8% | estimated ± 9.7 pp, low confidence |
| Terminal-Bench 3.0 | 4.0% | estimated ± 9.1 pp, low confidence |
| FrontierMath v2 (Tiers 1-3) | 41.1% | estimated ± 8.8 pp, medium confidence |
| Terminal-Bench 4.1 | 0.0% | estimated ± 10.9 pp, low confidence |
| Vals SWE-bench | 50.3% | estimated ± 6.6 pp, low confidence |
| Terminal-Bench 2.1 (Vals AI) | 47.0% | estimated ± 7.3 pp, low confidence |
| DeepSWE | 14.9% | estimated ± 12.7 pp, low confidence |
| HLE w/o tools | 23.7% | estimated ± 4.3 pp, high confidence |
| ARC-AGI-3 | 0.0% | estimated ± 8.6 pp, low confidence |
| Vibe Code Bench | 0.0% | estimated ± 8.4 pp, low confidence |
| CharXiv | 74.0% | estimated ± 3.3 pp, high confidence |
| CharXiv w/o tools | 72.2% | estimated ± 3.8 pp, medium confidence |
| AI2D_TEST | 88.7% | estimated ± 1.4 pp, medium confidence |
| C-Eval | 92.0% | estimated ± 1.2 pp, high confidence |
| CountBench | 93.4% | estimated ± 3.6 pp, high confidence |
| DynaMath | 81.4% | estimated ± 1.5 pp, medium confidence |
| ERQA | 55.7% | estimated ± 4.2 pp, high confidence |
| HMMT Feb 2025 | 96.1% | estimated ± 1.9 pp, low confidence |
| HMMT Feb 2026 | 88.6% | estimated ± 4.0 pp, high confidence |
| HMMT Nov 2025 | 93.5% | estimated ± 3.2 pp, high confidence |
| MathVision | 76.8% | estimated ± 9.1 pp, medium confidence |
| MLVU (M-Avg) | 79.0% | estimated ± 4.2 pp, medium confidence |
| MMAnswerBench | 84.2% | estimated ± 1.3 pp, high confidence |
| MMLU-Redux | 89.1% | measured |
| MMMU-Pro | 44.6% | estimated ± 7.6 pp, medium confidence |
| NL2Repo | 44.4% | estimated ± 3.4 pp, high confidence |
| OmniDocBench 1.5 | 87.9% | estimated ± 1.8 pp, medium confidence |
| RealWorldQA | 80.1% | measured |
| ScreenSpot Pro | 58.9% | estimated ± 4.8 pp, medium confidence |
| SimpleVQA | 56.9% | estimated ± 12.3 pp, low confidence |
| SuperGPQA | 58.3% | estimated ± 4.1 pp, high confidence |
| V* | 81.6% | estimated ± 10.1 pp, low confidence |
| Video-MME (with subtitle) | 79.4% | estimated ± 4.6 pp, high confidence |
| VideoMMMU | 44.7% | estimated ± 9.6 pp, low confidence |
| We-Math | 77.3% | estimated ± 5.5 pp, low confidence |
| MedXpertQA (Text) | 43.2% | estimated ± 4.5 pp, low confidence |
| MMLU-Pro (Arcee) | 85.4% | estimated ± 3.4 pp, medium confidence |
| React Native Evals | 71.0% | estimated ± 2.7 pp, medium confidence |
| SWE-Rebench | 57.0% | estimated ± 3.6 pp, low confidence |
| SimpleQA | 38.6% | estimated ± 13.4 pp, low confidence |
| LiveCodeBench | 66.6% | estimated ± 14.6 pp, low confidence |
| IMOAnswerBench | 74.5% | estimated ± 10.7 pp, low confidence |
| MRCRv2 | 85.2% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 61.3% | estimated ± 2.6 pp, low confidence |
| RefCOCO (avg) | 89.3% | estimated ± 3.1 pp, medium confidence |
| OCRBench V2 | 56.9% | measured |