Other · model
Inkling 2 Small benchmark scores
As of 2026-11-07, Inkling 2 Small (Other) has measured scores on 37 benchmarks and estimated scores on 125 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 91.1% | estimated ± 2.7 pp, high confidence |
| AA-HLE | 44.1% | measured |
| AA-SciCode | 52.3% | measured |
| AIME26 | 96.9% | estimated ± 1.3 pp, medium confidence |
| CritPt | 19.2% | estimated ± 2.5 pp, high confidence |
| GDPval-AA | 61.0% | estimated ± 6.2 pp, medium confidence |
| GPQA Diamond | 93.0% | measured |
| HLE | 53.2% | measured |
| AA-LCR | 83.3% | measured |
| LiveCodeBench v6 | 92.8% | estimated ± 4.9 pp, medium confidence |
| SciCode | 52.1% | estimated ± 3.1 pp, high confidence |
| Terminal-Bench 2.1 | 85.1% | measured |
| BrowseComp | 90.2% | measured |
| HLE w/ tools | 56.4% | estimated ± 5.2 pp, medium confidence |
| LongBench v2 | 70.6% | measured |
| VITA-Bench | 47.1% | estimated ± 6.7 pp, low confidence |
| τ²-bench results | 93.8% | estimated ± 11.5 pp, low confidence |
| AA Coding Index | 76.8% | estimated ± 2.5 pp, high confidence |
| AA-MMMU-Pro | 81.6% | measured |
| APEX-Agents | 37.9% | estimated ± 0.6 pp, medium confidence |
| APEX-Agents-AA | 43.7% | estimated ± 3.6 pp, low confidence |
| BioMysteryBench (human-difficult) | 44.1% | estimated ± 8.1 pp, low confidence |
| DeepSearchQA | 90.9% | measured |
| SWE-bench Verified | 82.2% | measured |
| AutomationBench | 30.9% | estimated ± 8.2 pp, low confidence |
| BFCL v4 | 77.3% | estimated ± 4.7 pp, low confidence |
| CyberGym | 81.1% | estimated ± 3.1 pp, low confidence |
| JobBench | 70.5% | measured |
| SWE-bench Pro | 60.9% | estimated ± 4.5 pp, high confidence |
| WideResearch | 85.7% | estimated ± 1.2 pp, low confidence |
| AA Agentic Index | 42.9% | estimated ± 5.8 pp, medium confidence |
| MMLU-Pro | 85.7% | estimated ± 2.5 pp, low confidence |
| Artificial Analysis Intelligence Index | 30.6% | measured |
| GPQA | 93.0% | measured |
| MMLU | 92.1% | estimated ± 1.0 pp, high confidence |
| AA-Omniscience Accuracy | 33.8% | measured |
| MMMLU | 90.8% | estimated ± 1.1 pp, low confidence |
| MMMU | 87.1% | estimated ± 0.8 pp, medium confidence |
| AA MMLU-Pro | 87.1% | estimated ± 0.6 pp, medium confidence |
| AIME 2025 | 98.8% | estimated ± 3.8 pp, low confidence |
| Gert Labs | 100.0% | estimated ± 7.3 pp, low confidence |
| AA-AnalystAgent | 51.2% | measured |
| AA EnterpriseOps-Gym | 46.8% | measured |
| AA Harvey LAB v1.0 | 93.0% | estimated ± 9.8 pp, low confidence |
| AA Tau3 Banking | 40.4% | measured |
| AA Terminal-Bench 2.1 | 81.3% | measured |
| ApprenticeBench | 21.1% | estimated ± 5.6 pp, low confidence |
| ARC-AGI-1 | 93.3% | estimated ± 2.7 pp, high confidence |
| ARC-AGI-2 | 71.4% | estimated ± 10.1 pp, low confidence |
| CursorBench 3.1 | 49.7% | estimated ± 3.4 pp, medium confidence |
| CursorBench 3.2 | 57.4% | estimated ± 5.3 pp, medium confidence |
| Terminal-Bench 3.0 | 18.5% | estimated ± 8.6 pp, low confidence |
| FrontierCode 1.1 Main | 32.5% | estimated ± 2.7 pp, medium confidence |
| FrontierMath v2 (Tier 4) | 33.4% | estimated ± 4.3 pp, medium confidence |
| FrontierMath v2 (Tiers 1-3) | 55.6% | estimated ± 9.0 pp, medium confidence |
| FrontierSWE v2 | 22.7% | estimated ± 15.0 pp, low confidence |
| MazeBench | 0.1% | estimated ± 5.1 pp, medium confidence |
| OfficeQA Pro | 66.7% | estimated ± 2.5 pp, medium confidence |
| OSWorld-Verified | 81.3% | estimated ± 4.4 pp, medium confidence |
| ParseBench | 71.1% | estimated ± 0.5 pp, medium confidence |
| Vals GPQA Diamond | 92.3% | estimated ± 2.0 pp, high confidence |
| Vals LiveCodeBench | 84.7% | estimated ± 9.9 pp, low confidence |
| Vals MMLU-Pro | 88.2% | estimated ± 1.2 pp, high confidence |
| Vals SWE-bench | 79.4% | estimated ± 6.8 pp, medium confidence |
| Terminal-Bench 2.1 (Vals AI) | 69.9% | estimated ± 6.1 pp, low confidence |
| AA AutomationBench | 45.6% | measured |
| CWE-Bench-AA | 55.3% | estimated ± 6.4 pp, medium confidence |
| AA Cyber Index | 42.4% | estimated ± 13.5 pp, low confidence |
| DeepsecBench-AA | 8.0% | estimated ± 5.5 pp, low confidence |
| GDP.pdf | 25.2% | measured |
| AA ITBench | 50.1% | estimated ± 4.1 pp, high confidence |
| AA Terminal-Bench 4.0 | 13.1% | measured |
| Bug Hunt Bench | 15.3% | estimated ± 3.2 pp, low confidence |
| CursorBench 4.0 | 33.3% | estimated ± 3.5 pp, medium confidence |
| CWE-bench v1 | 34.3% | estimated ± 4.2 pp, low confidence |
| DeepSWE | 72.1% | measured |
| HLE w/o tools | 44.1% | measured |
| OSWorld 2.0 | 52.8% | estimated ± 9.2 pp, low confidence |
| OSWorld 2.1 | 77.2% | estimated ± 2.8 pp, medium confidence |
| PostTrainBench v1.1 | 40.1% | estimated ± 4.5 pp, high confidence |
| SWE Multilingual | 81.2% | measured |
| Terminal-Bench 4.0 | 17.7% | estimated ± 5.8 pp, low confidence |
| Toolathlon-Verified | 77.4% | estimated ± 1.6 pp, medium confidence |
| OSWorld 2.1 Offline | 71.0% | estimated ± 9.2 pp, low confidence |
| Vibe Code Bench | 59.3% | estimated ± 8.1 pp, medium confidence |
| ARC-AGI-3 | 0.3% | estimated ± 2.0 pp, high confidence |
| ARC-AGI-3 Provider Adapter | 4.2% | estimated ± 7.5 pp, medium confidence |
| Hermes Bench | 67.9% | estimated ± 3.9 pp, high confidence |
| Hermes Index | 33.6% | estimated ± 8.8 pp, medium confidence |
| CharXiv | 87.7% | measured |
| CharXiv w/o tools | 85.6% | measured |
| ExploitGym | 28.0% | estimated ± 9.2 pp, low confidence |
| AI2D_TEST | 93.5% | estimated ± 0.9 pp, low confidence |
| C-Eval | 94.7% | estimated ± 1.0 pp, low confidence |
| Claw-Eval | 72.0% | estimated ± 6.3 pp, low confidence |
| CountBench | 97.3% | estimated ± 1.3 pp, low confidence |
| DeepPlanning | 52.4% | estimated ± 13.1 pp, low confidence |
| DynaMath | 87.6% | estimated ± 0.8 pp, low confidence |
| ERQA | 72.6% | estimated ± 2.9 pp, medium confidence |
| HMMT Feb 2026 | 95.8% | measured |
| HMMT Nov 2025 | 96.5% | estimated ± 2.8 pp, medium confidence |
| INCLUDE | 86.8% | estimated ± 2.0 pp, medium confidence |
| MathVision | 92.8% | measured |
| MCP Atlas | 85.0% | measured |
| MLVU (M-Avg) | 88.5% | estimated ± 1.5 pp, high confidence |
| MMAnswerBench | 88.1% | estimated ± 2.8 pp, low confidence |
| MMLU-ProX | 87.2% | estimated ± 1.3 pp, medium confidence |
| MMLU-Redux | 95.2% | estimated ± 0.8 pp, medium confidence |
| MMMU-Pro | 81.6% | measured |
| NL2Repo | 50.4% | estimated ± 6.2 pp, medium confidence |
| OmniDocBench 1.5 | 91.2% | estimated ± 2.2 pp, high confidence |
| QwenClawBench | 70.2% | estimated ± 2.6 pp, medium confidence |
| RealWorldQA | 87.8% | estimated ± 1.0 pp, medium confidence |
| ScreenSpot Pro | 83.8% | measured |
| SimpleVQA | 76.2% | estimated ± 7.3 pp, medium confidence |
| SuperGPQA | 92.0% | estimated ± 6.9 pp, low confidence |
| Toolathlon | 61.0% | estimated ± 4.2 pp, high confidence |
| V* | 96.0% | estimated ± 6.5 pp, low confidence |
| Video-MME (w/o subtitle) | 86.3% | estimated ± 1.4 pp, low confidence |
| Video-MME (with subtitle) | 90.1% | estimated ± 1.9 pp, high confidence |
| VideoMMMU | 86.7% | estimated ± 1.5 pp, medium confidence |
| We-Math | 87.5% | estimated ± 2.1 pp, low confidence |
| HealthBench Hard | 35.1% | estimated ± 5.1 pp, low confidence |
| MedXpertQA (MM) | 77.9% | estimated ± 2.4 pp, high confidence |
| MedXpertQA (Text) | 63.9% | estimated ± 4.5 pp, medium confidence |
| MMLU-Pro (Arcee) | 87.9% | estimated ± 3.4 pp, low confidence |
| React Native Evals | 88.7% | estimated ± 2.3 pp, low confidence |
| ResearchClawBench | 19.6% | estimated ± 1.8 pp, low confidence |
| SWE-Rebench | 61.8% | estimated ± 3.6 pp, medium confidence |
| Finance Agent v2 | 57.3% | estimated ± 4.2 pp, medium confidence |
| BioMysteryBench (human-solvable) | 87.5% | estimated ± 1.4 pp, low confidence |
| Chartography (tools) | 80.4% | measured |
| DRACO | 83.7% | estimated ± 4.5 pp, medium confidence |
| HealthBench (raw) | 52.4% | estimated ± 6.0 pp, medium confidence |
| HealthBench (length-adjusted) | 55.1% | estimated ± 3.4 pp, medium confidence |
| HealthBench Professional | 58.4% | estimated ± 4.5 pp, medium confidence |
| HealthBench Professional (raw) | 62.8% | estimated ± 4.5 pp, high confidence |
| HLE-Verified | 53.8% | estimated ± 6.8 pp, low confidence |
| LABBench2 | 81.5% | estimated ± 2.4 pp, medium confidence |
| ScarfBench | 67.6% | estimated ± 3.8 pp, medium confidence |
| STEMBench | 2.4% | estimated ± 0.2 pp, medium confidence |
| SimpleQA | 37.1% | estimated ± 4.3 pp, low confidence |
| LiveCodeBench | 80.6% | estimated ± 14.6 pp, low confidence |
| Agents' Last Exam | 45.7% | estimated ± 11.0 pp, low confidence |
| BabyVision w/ Python | 93.7% | measured |
| OpenHarmony Bench | 58.8% | estimated ± 2.4 pp, high confidence |
| IMOAnswerBench | 86.3% | estimated ± 5.6 pp, medium confidence |
| BabyVision | 68.6% | measured |
| MMVU | 79.3% | estimated ± 2.7 pp, medium confidence |
| PerceptionBench | 64.7% | measured |
| ZeroBench | 19.7% | estimated ± 4.7 pp, low confidence |
| MRCRv2 | 85.9% | estimated ± 11.3 pp, low confidence |
| LVBench | 80.9% | estimated ± 8.2 pp, low confidence |
| MMLongBench-Doc | 61.5% | estimated ± 2.6 pp, low confidence |
| SWE-Marathon | 38.8% | measured |
| BioBasicsBench | 33.4% | estimated ± 2.2 pp, medium confidence |
| LifeVal | 79.7% | estimated ± 1.6 pp, medium confidence |
| MMMU-Pro w/ Python | 84.2% | estimated ± 2.0 pp, high confidence |
| OCRBench V2 | 73.4% | estimated ± 4.4 pp, medium confidence |
| MathVision w/ Python | 96.6% | estimated ± 0.5 pp, medium confidence |
| PinchBench | 90.7% | estimated ± 3.6 pp, low confidence |
| SWE-Atlas Codebase QnA | 61.2% | estimated ± 5.4 pp, low confidence |