Raw values
Scores are shown as published, without price- or context-based adjustments.
Benchmark registry
Browse raw benchmark results by version and metric. Each ranking stays tied to its source and evaluation date.
Scores are shown as published, without price- or context-based adjustments.
Different benchmark versions are listed and ranked independently.
Coverage, source count, and the latest evaluation date stay visible.
Available data
| Benchmark | Version | Metric | Models | Sources | Latest evaluation | Ranking |
|---|---|---|---|---|---|---|
| Intelligence Index | Unversioned | index | 334 | 1 | Not documented | View ranking |
| Coding Index | Unversioned | index | 214 | 1 | Not documented | View ranking |
| Agentic Index | Unversioned | index | 201 | 1 | Not documented | View ranking |
| Design Arena: website | Unversioned | elo | 200 | 1 | Not documented | View ranking |
| Design Arena: gamedev | Unversioned | elo | 191 | 1 | Not documented | View ranking |
| Design Arena: codecategories | Unversioned | elo | 188 | 1 | Not documented | View ranking |
| Design Arena: dataviz | Unversioned | elo | 183 | 1 | Not documented | View ranking |
| Design Arena: uicomponent | Unversioned | elo | 182 | 1 | Not documented | View ranking |
| Design Arena: 3d | Unversioned | elo | 176 | 1 | Not documented | View ranking |
| Design Arena: svg | Unversioned | elo | 128 | 1 | Not documented | View ranking |
| Design Arena: asciiart | Unversioned | elo | 108 | 1 | Not documented | View ranking |
| Design Arena: webapps | Unversioned | elo | 80 | 1 | Not documented | View ranking |
| Design Arena: mobileapps | Unversioned | elo | 77 | 1 | Not documented | View ranking |
| Design Arena: fullstack | Unversioned | elo | 69 | 1 | Not documented | View ranking |
| Design Arena: androidnative | Unversioned | elo | 61 | 1 | Not documented | View ranking |
| Design Arena: godotgamedev | Unversioned | elo | 55 | 1 | Not documented | View ranking |
| Design Arena: python-pptxslides | Unversioned | elo | 48 | 1 | Not documented | View ranking |
| SWE-Bench Pro | Unversioned | resolve rate | 45 | 23 | 2026-08-10 | View ranking |
| Design Arena: agenticgamedev | Unversioned | elo | 44 | 1 | Not documented | View ranking |
| Design Arena: htmlslides | Unversioned | elo | 43 | 1 | Not documented | View ranking |
| SWE-Bench Verified | Unversioned | resolved | 36 | 30 | 2026-07-24 | View ranking |
| Design Arena: blender3d | Unversioned | elo | 35 | 1 | Not documented | View ranking |
| Aider Polyglot | Unversioned | percent correct | 31 | 1 | 2025-10-03 | View ranking |
| Design Arena: spreadsheet | Unversioned | elo | 31 | 1 | Not documented | View ranking |
| SciCode | Unversioned | percent correct | 27 | 26 | 2026-06-15 | View ranking |
| Artificial Analysis Coding Index | Unversioned | index | 26 | 25 | 2026-06-15 | View ranking |
| Humanity's Last Exam | Unversioned | accuracy | 23 | 16 | 2026-09-03 | View ranking |
| Terminal-Bench Hard | Unversioned | success rate | 22 | 21 | 2026-06-15 | View ranking |
| Terminal-Bench | 2.1 | success rate | 20 | 13 | 2026-08-10 | View ranking |
| GPQA Diamond | Unversioned | accuracy | 19 | 12 | 2026-09-03 | View ranking |
| Design Arena: agentichtmlslides | Unversioned | elo | 17 | 1 | Not documented | View ranking |
| Design Arena: agenticslides | Unversioned | elo | 17 | 1 | Not documented | View ranking |
| Design Arena: agenticslides(html) | Unversioned | elo | 17 | 1 | Not documented | View ranking |
| Design Arena: agenticslides(python-pptx) | Unversioned | elo | 17 | 1 | Not documented | View ranking |
| OSWorld-Verified | Unversioned | success rate | 17 | 11 | 2026-08-10 | View ranking |
| Agents' Last Exam | Unversioned | score | 16 | 12 | 2026-09-03 | View ranking |
| Terminal-Bench | 2.1 | pass@1 | 16 | 7 | 2026-09-02 | View ranking |
| BrowseComp | Unversioned | accuracy | 15 | 10 | 2026-09-03 | View ranking |
| Design Arena: pptxslides | Unversioned | elo | 15 | 1 | Not documented | View ranking |
| DeepSWE | 1.1 | score | 13 | 9 | 2026-10-16 | View ranking |
| Humanity's Last Exam | Unversioned | score | 12 | 12 | 2026-10-16 | View ranking |
| SWE-Atlas Codebase QnA | Unversioned | score | 12 | 3 | 2026-10-16 | View ranking |
| DeepSWE | 1.1 | resolve rate | 11 | 9 | 2026-09-03 | View ranking |
| SWE-Atlas Codebase QnA | Unversioned | pass@1 | 11 | 2 | 2026-09-02 | View ranking |
| SWE-Atlas Refactoring | Unversioned | score | 11 | 1 | Not documented | View ranking |
| MMMU Pro | Unversioned | accuracy | 10 | 5 | 2026-08-03 | View ranking |
| NL2Repo | Unversioned | score | 10 | 10 | 2026-05-19 | View ranking |
| SWE-Atlas Test Writing | Unversioned | score | 10 | 1 | Not documented | View ranking |
| Terminal-Bench | 4.0 | score | 10 | 7 | 2026-10-07 | View ranking |
| Artificial Analysis Coding Agent Index | Unversioned | average pass@1 | 9 | 1 | Not documented | View ranking |
| SWE-Bench Pro | Unversioned | pass@1 | 9 | 1 | Not documented | View ranking |
| Terminal-Bench | 2.1 | score | 9 | 7 | 2026-10-16 | View ranking |
| GDPval-AA | 2 | Elo | 8 | 7 | 2026-10-15 | View ranking |
| GPQA Diamond | Unversioned | score | 8 | 7 | 2026-10-16 | View ranking |
| JobBench | Unversioned | score | 8 | 6 | 2026-10-15 | View ranking |
| MCP Atlas | Unversioned | success rate | 8 | 7 | 2026-08-10 | View ranking |
| CharXiv Reasoning | Unversioned | accuracy | 7 | 6 | 2026-08-10 | View ranking |
| SWE-Bench Multilingual | Unversioned | resolve rate | 7 | 7 | 2026-07-24 | View ranking |
| Toolathlon-Verified | Unversioned | score | 7 | 5 | 2026-10-15 | View ranking |
| ARC-AGI-2 | Unversioned | accuracy | 6 | 3 | 2026-07-24 | View ranking |
| CyberGym | Unversioned | score | 6 | 4 | 2026-07-31 | View ranking |
| GDPval | Unversioned | wins or ties | 6 | 3 | 2026-06-04 | View ranking |
| GDPval-AA | Unversioned | Elo | 6 | 5 | 2026-06-09 | View ranking |
| MCP Atlas | Unversioned | score | 6 | 5 | 2026-10-16 | View ranking |
| SciCode | Unversioned | score | 6 | 5 | 2026-10-16 | View ranking |
| Terminal-Bench | 2.1 | accuracy | 6 | 6 | 2026-08-13 | View ranking |
| Toolathlon | Unversioned | success rate | 6 | 4 | 2026-07-09 | View ranking |
| AutomationBench | Unversioned | pass@1 | 5 | 4 | 2026-09-02 | View ranking |
| AutomationBench | Unversioned | success rate | 5 | 5 | 2026-09-03 | View ranking |
| AutomationBench | 1.0.6 | score | 5 | 3 | 2026-09-22 | View ranking |
| CharXiv Reasoning | Unversioned | score | 5 | 4 | Not documented | View ranking |
| DeepSWE | 1.1 | pass@1 | 5 | 4 | 2026-10-12 | View ranking |
| Design Arena: graphicdesign | Unversioned | elo | 5 | 1 | Not documented | View ranking |
| Design Arena: image | Unversioned | elo | 5 | 1 | Not documented | View ranking |
| Design Arena: logo | Unversioned | elo | 5 | 1 | Not documented | View ranking |
| GDPval-AA | 2.1 | Elo | 5 | 4 | 2026-10-07 | View ranking |
| IFBench | Unversioned | score | 5 | 5 | 2026-10-16 | View ranking |
| Terminal-Bench | 2.0 | success rate | 5 | 4 | 2026-07-02 | View ranking |
| Chartography | Unversioned | score | 4 | 4 | 2026-10-07 | View ranking |
| CursorBench | 4.0 | score | 4 | 3 | 2026-09-28 | View ranking |
| FrontierCode | 1.1 | score | 4 | 4 | 2026-10-07 | View ranking |
| FrontierMath | Unversioned | accuracy | 4 | 1 | 2026-04-23 | View ranking |
| FrontierMath | v2 | accuracy | 4 | 2 | 2026-09-03 | View ranking |
| FrontierSWE | Unversioned | dominance score | 4 | 4 | 2026-08-14 | View ranking |
| GDPval-AA | v2 | Elo | 4 | 4 | 2026-07-24 | View ranking |
| OSWorld | 2.0 | partial score | 4 | 4 | 2026-09-22 | View ranking |
| OSWorld | 2.0 | success rate | 4 | 2 | 2026-07-24 | View ranking |
| OSWorld-Verified | Unversioned | score | 4 | 3 | 2026-10-15 | View ranking |
| ProgramBench | Unversioned | score | 4 | 2 | Not documented | View ranking |
| Terminal-Bench | 2.0 | accuracy | 4 | 3 | 2026-03-17 | View ranking |
| ZeroBench | Unversioned | pass@5 | 4 | 4 | 2026-07-16 | View ranking |
| AA-Briefcase | 1.1 | Elo | 3 | 2 | 2026-10-07 | View ranking |
| AA-LCR | Unversioned | score | 3 | 3 | 2026-10-16 | View ranking |
| Artificial Analysis Coding Agent Index | 1.1 | index score | 3 | 1 | 2026-07-09 | View ranking |
| Artificial Analysis Intelligence Index | 4.1 | index score | 3 | 1 | 2026-07-09 | View ranking |
| AutomationBench | Unversioned | score | 3 | 3 | 2026-10-16 | View ranking |
| BabyVision | Unversioned | score | 3 | 3 | 2026-10-15 | View ranking |
| BrowseComp | Unversioned | score | 3 | 3 | 2026-10-16 | View ranking |
| Claw-eval | Unversioned | percent | 3 | 3 | Not documented | View ranking |
| CoWorkBench | Unversioned | score | 3 | 3 | Not documented | View ranking |
| Codeforces | Unversioned | rating | 3 | 3 | Not documented | View ranking |
| DSBench-Hard | Unversioned | score | 3 | 3 | 2026-07-31 | View ranking |
| DeepSWE | Unversioned | score | 3 | 3 | Not documented | View ranking |
| DeepSWE | 1.1 | resolved | 3 | 3 | Not documented | View ranking |
| DeepSearchQA | Unversioned | F1 | 3 | 2 | 2026-09-02 | View ranking |
| ExploitBench | Unversioned | score | 3 | 1 | Not documented | View ranking |
| ExploitGym | Unversioned | score | 3 | 1 | Not documented | View ranking |
| Finance Agent | v2 | accuracy | 3 | 2 | 2026-10-12 | View ranking |
| GDM-MRCR | v2 | accuracy | 3 | 3 | 2026-08-13 | View ranking |
| GPQA Diamond | Unversioned | pass@1 | 3 | 3 | Not documented | View ranking |
| HealthBench | Unversioned | score | 3 | 2 | 2026-09-21 | View ranking |
| Humanity's Last Exam | Unversioned | pass@1 | 3 | 3 | Not documented | View ranking |
| IFBench | Unversioned | accuracy | 3 | 3 | 2026-06-04 | View ranking |
| LiveCodeBench | Unversioned | score | 3 | 2 | 2026-10-16 | View ranking |
| MathArena Apex | Unversioned | pass@1 | 3 | 3 | Not documented | View ranking |
| MiMo Code Bench | Unversioned | score | 3 | 1 | Not documented | View ranking |
| MiMo Cyber Bench | Unversioned | score | 3 | 1 | Not documented | View ranking |
| NL2Repo | Unversioned | percent | 3 | 3 | Not documented | View ranking |
| NL2Repo | Unversioned | resolve rate | 3 | 3 | 2026-08-03 | View ranking |
| Program Bench | Unversioned | score | 3 | 3 | 2026-07-16 | View ranking |
| SEC-Bench Pro | Unversioned | score | 3 | 1 | Not documented | View ranking |
| SWE-Bench Multilingual | Unversioned | percent resolved | 3 | 3 | Not documented | View ranking |
| SWE-Bench Multilingual | Unversioned | score | 3 | 3 | 2026-10-16 | View ranking |
| SWE-Bench Pro | Unversioned | percent resolved | 3 | 3 | Not documented | View ranking |
| SWE-Bench Pro | Unversioned | resolved | 3 | 3 | Not documented | View ranking |
| SWE-Bench Pro | Unversioned | score | 3 | 3 | Not documented | View ranking |
| SWE-Bench Verified | Unversioned | percent resolved | 3 | 3 | Not documented | View ranking |
| Terminal-Bench 2.1 | Unversioned | percent | 3 | 3 | Not documented | View ranking |
| Toolathlon | Unversioned | score | 3 | 2 | 2026-03-17 | View ranking |
| Toolathlon-Verified | Unversioned | pass@1 | 3 | 3 | Not documented | View ranking |
| τ3 Banking | Unversioned | score | 3 | 2 | 2026-10-16 | View ranking |
| τ²-Bench Telecom | Unversioned | success rate | 3 | 3 | 2026-06-09 | View ranking |
| AA-Briefcase | Unversioned | Elo | 2 | 2 | 2026-07-24 | View ranking |
| AA-Omniscience | Unversioned | score | 2 | 2 | 2026-10-15 | View ranking |
| ARC-AGI-1 | Unversioned | accuracy | 2 | 2 | 2026-07-24 | View ranking |
| ARC-AGI-3 | Unversioned | RHAE | 2 | 2 | 2026-09-03 | View ranking |
| Agentic IF Index | Unversioned | score | 2 | 1 | 2026-09-02 | View ranking |
| AndroidWorld | Unversioned | score | 2 | 2 | Not documented | View ranking |
| Artificial Analysis Intelligence Index | Unversioned | index score | 2 | 2 | 2026-10-15 | View ranking |
| Artificial Analysis Intelligence Index | 4.3 | index score | 2 | 1 | 2026-10-12 | View ranking |
| AutomationBench | Unversioned | pass rate | 2 | 2 | 2026-09-22 | View ranking |
| AutomationBench | Unversioned | strict pass@1 | 2 | 1 | 2026-10-12 | View ranking |
| AutomationBench | 1.0.6 | pass@1 | 2 | 2 | Not documented | View ranking |
| BrowseComp | Unversioned | pass@1 | 2 | 2 | Not documented | View ranking |
| CTI-REALM | Unversioned | score | 2 | 1 | Not documented | View ranking |
| Chinese SimpleQA | Unversioned | pass@1 | 2 | 2 | Not documented | View ranking |
| ClawEval-MM | Unversioned | average score | 2 | 2 | Not documented | View ranking |
| CorpusQA | Unversioned | accuracy | 2 | 2 | Not documented | View ranking |
| CyberGym | Unversioned | pass@1 | 2 | 2 | Not documented | View ranking |
| DSBench-FullStack | Unversioned | score | 2 | 2 | 2026-07-31 | View ranking |
| DeepSWE | Unversioned | resolve rate | 2 | 2 | 2026-07-31 | View ranking |
| DeepSearchQA | Unversioned | score | 2 | 2 | 2026-10-16 | View ranking |
| EEBench | Unversioned | score | 2 | 1 | 2026-09-21 | View ranking |
| ERQA | Unversioned | score | 2 | 2 | Not documented | View ranking |
| FrontierCode | Unversioned | pass rate | 2 | 1 | 2026-06-09 | View ranking |
| GDP.pdf | Unversioned | score | 2 | 1 | 2026-10-12 | View ranking |
| GDPval | Unversioned | Elo | 2 | 1 | 2026-09-21 | View ranking |
| GeneBench | Unversioned | accuracy | 2 | 1 | 2026-04-23 | View ranking |
| Graphwalks | Unversioned | accuracy | 2 | 1 | 2026-03-17 | View ranking |
| HMMT | Unversioned | pass@1 | 2 | 2 | Not documented | View ranking |
| Harvey Legal Agent Benchmark | Unversioned | score | 2 | 1 | 2026-09-21 | View ranking |
| Harvey Legal Agent Benchmark | 1.1 | all-pass rate | 2 | 1 | 2026-10-12 | View ranking |
| HealthBench Professional | Unversioned | length-normalized weighted rubric score | 2 | 1 | 2026-10-12 | View ranking |
| Humanity’s Last Exam | Unversioned | score | 2 | 1 | Not documented | View ranking |
| IMOAnswerBench | Unversioned | pass@1 | 2 | 2 | Not documented | View ranking |
| JobBench | Unversioned | mean reward | 2 | 1 | 2026-10-12 | View ranking |
| JobBench | Unversioned | mean rubric score | 2 | 1 | 2026-09-02 | View ranking |
| LiveCodeBench | Unversioned | pass@1 | 2 | 2 | Not documented | View ranking |
| LiveCodeBench | 6 | score | 2 | 2 | Not documented | View ranking |
| LiveCodeBench Pro | Unversioned | score | 2 | 1 | Not documented | View ranking |
| Long Context Reasoning | Unversioned | score | 2 | 1 | Not documented | View ranking |
| LongBench | 2 | score | 2 | 2 | 2026-10-16 | View ranking |
| MCP Atlas | Unversioned | pass@1 | 2 | 2 | Not documented | View ranking |
| MLE-Bench | Unversioned | average position score | 2 | 2 | 2026-07-21 | View ranking |
| MLS-Bench-Lite | Unversioned | score | 2 | 2 | 2026-08-03 | View ranking |
| MMLU-Pro | Unversioned | EM | 2 | 2 | Not documented | View ranking |
| MMLU-Pro | Unversioned | score | 2 | 2 | Not documented | View ranking |
| MRCR | Unversioned | MMR | 2 | 2 | Not documented | View ranking |
| MRCR | 2 | mean sequence-match ratio | 2 | 1 | 2026-09-02 | View ranking |
| MRCRv2 | Unversioned | score | 2 | 1 | Not documented | View ranking |
| MathArena Apex Shortlist | Unversioned | pass@1 | 2 | 2 | Not documented | View ranking |
| MathVision | Unversioned | score | 2 | 2 | Not documented | View ranking |
| MiMo VisualCoding | Unversioned | score | 2 | 1 | Not documented | View ranking |
| OSWorld | 2.0 | binary completion rate | 2 | 1 | 2026-09-02 | View ranking |
| OSWorld | 2.1 | binary completion rate | 2 | 1 | 2026-10-12 | View ranking |
| OSWorld | 2.1 | partial score | 2 | 2 | 2026-10-07 | View ranking |
| OmniDocBench | 1.5 | overall edit distance | 2 | 1 | 2026-03-17 | View ranking |
| OpenAI MRCR | v2 | accuracy | 2 | 1 | 2026-03-17 | View ranking |
| ProgramBench | Unversioned | share of rollouts passing at least 95% of tests | 2 | 1 | 2026-10-12 | View ranking |
| QwenSWEBench | Unversioned | avg@3 | 2 | 2 | Not documented | View ranking |
| RealWorldQA | Unversioned | score | 2 | 2 | Not documented | View ranking |
| RecreationBench | Unversioned | score | 2 | 2 | Not documented | View ranking |
| SWE Bench Pro | Unversioned | score | 2 | 1 | Not documented | View ranking |
| SWE-Bench Multilingual | Unversioned | resolved | 2 | 2 | Not documented | View ranking |
| SWE-Marathon | Unversioned | pass@1 | 2 | 1 | 2026-10-12 | View ranking |
| SimpleQA-Verified | Unversioned | pass@1 | 2 | 2 | Not documented | View ranking |
| SpreadsheetBench | 2 | score | 2 | 2 | 2026-07-16 | View ranking |
| Terminal Bench 2.1 | Unversioned | score | 2 | 1 | Not documented | View ranking |
| Terminal-Bench | 4.0 | accuracy | 2 | 1 | 2026-10-12 | View ranking |
| Terminal-Bench | 4.0 | pass@1 | 2 | 2 | Not documented | View ranking |
| Terminal-Bench-Science | 0.1 | score | 2 | 2 | 2026-09-22 | View ranking |
| Toolathlon | Unversioned | pass@1 | 2 | 2 | Not documented | View ranking |
| Vision2Web | Unversioned | score | 2 | 2 | Not documented | View ranking |
| tau3-bench | Unversioned | score | 2 | 2 | Not documented | View ranking |
| τ²-Bench Telecom | Unversioned | accuracy | 2 | 1 | 2026-03-17 | View ranking |
| $OneMillion-Bench | Unversioned | expert score | 1 | 1 | Not documented | View ranking |
| AA-Omniscience | Unversioned | index | 1 | 1 | 2026-10-16 | View ranking |
| AIME | 2026 | accuracy | 1 | 1 | 2026-06-16 | View ranking |
| AIME 2026 | Unversioned | accuracy | 1 | 1 | 2026-08-10 | View ranking |
| AIME 2026 | Unversioned | score | 1 | 1 | 2026-10-16 | View ranking |
| AIME26 | Unversioned | score | 1 | 1 | Not documented | View ranking |
| APEX-Agents | Unversioned | pass@1 | 1 | 1 | Not documented | View ranking |
| APEX-Agents | Unversioned | score | 1 | 1 | Not documented | View ranking |
| Agents' Last Exam | Unversioned | pass@1 | 1 | 1 | Not documented | View ranking |
| Agents' Last Exam | 1 | score | 1 | 1 | 2026-09-22 | View ranking |
| AndroidBench | Unversioned | avg@3 | 1 | 1 | Not documented | View ranking |
| Apex Bench | 1.1 | score | 1 | 1 | Not documented | View ranking |
| ApexBench | Unversioned | pass@1 | 1 | 1 | Not documented | View ranking |
| Artificial Analysis Coding Agent Index | 1.4 | index score | 1 | 1 | 2026-09-03 | View ranking |
| Artificial Analysis Coding Index | Unversioned | index score | 1 | 1 | 2026-06-09 | View ranking |
| Artificial Analysis Intelligence Index | 4.0 | index score | 1 | 1 | 2026-04-30 | View ranking |
| Artificial Analysis Intelligence Index | 4.1.1 | index score | 1 | 1 | 2026-09-03 | View ranking |
| ArxivMath | Unversioned | score | 1 | 1 | Not documented | View ranking |
| AutomationBench | Unversioned | accuracy | 1 | 1 | 2026-08-13 | View ranking |
| BabyVision | Unversioned | accuracy | 1 | 1 | 2026-07-09 | View ranking |
| BabyVision | Unversioned | pass@1 | 1 | 1 | Not documented | View ranking |
| BenchCAD | Unversioned | geometric overlap | 1 | 1 | 2026-09-03 | View ranking |
| BioMysteryBench | Unversioned | pass@1 | 1 | 1 | Not documented | View ranking |
| CL-bench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| CL-bench-life | Unversioned | score | 1 | 1 | Not documented | View ranking |
| CMT Benchmark | Unversioned | score | 1 | 1 | Not documented | View ranking |
| CharXiv Reasoning | Unversioned | pass@1 | 1 | 1 | Not documented | View ranking |
| Chartography | Unversioned | pass@1 | 1 | 1 | Not documented | View ranking |
| Claw-Eval | Unversioned | score | 1 | 1 | Not documented | View ranking |
| ClawEval | 1.1 | pass^3 | 1 | 1 | 2026-05-29 | View ranking |
| ClawEval | 20260325 | pass@3 | 1 | 1 | Not documented | View ranking |
| CritPt | Unversioned | accuracy | 1 | 1 | 2026-06-16 | View ranking |
| CritPt | Unversioned | score | 1 | 1 | 2026-10-16 | View ranking |
| CursorBench | 3.2.0 | score | 1 | 1 | 2026-09-01 | View ranking |
| DailyOmni | Unversioned | score | 1 | 1 | Not documented | View ranking |
| DeepSWE | 1.0 | resolve rate | 1 | 1 | 2026-07-08 | View ranking |
| DeepSearch QA | Unversioned | score | 1 | 1 | 2026-08-10 | View ranking |
| Design Arena: imageediting | Unversioned | elo | 1 | 1 | Not documented | View ranking |
| ExploitBench | Unversioned | average coverage | 1 | 1 | Not documented | View ranking |
| ExploitBench | Unversioned | success rate | 1 | 1 | 2026-09-03 | View ranking |
| ExploitGym | Unversioned | pass@1 | 1 | 1 | Not documented | View ranking |
| ExploitGym | Unversioned | tasks solved | 1 | 1 | Not documented | View ranking |
| FORTE | Unversioned | success rate | 1 | 1 | 2026-06-30 | View ranking |
| FairPoliticsQA | Unversioned | score | 1 | 1 | Not documented | View ranking |
| FinCRAFT | Unversioned | score | 1 | 1 | Not documented | View ranking |
| FinFIRST | Unversioned | score | 1 | 1 | Not documented | View ranking |
| FinSearchComp Verified | Unversioned | score | 1 | 1 | Not documented | View ranking |
| Finance Agent | 1.1 | strict-pass | 1 | 1 | Not documented | View ranking |
| Finance Agent | 2 | pass@1 | 1 | 1 | Not documented | View ranking |
| Finance Agent | 2 | strict-pass | 1 | 1 | Not documented | View ranking |
| FinanceAgent | 1.1 | accuracy | 1 | 1 | 2026-04-23 | View ranking |
| Frontier-Bench | v0.1 | mean reward | 1 | 1 | 2026-07-24 | View ranking |
| FrontierCode | 1.1 | mean@5 | 1 | 1 | 2026-07-24 | View ranking |
| FrontierCode | 1.1 Main | score | 1 | 1 | 2026-08-13 | View ranking |
| FrontierCode | v1 | pass rate | 1 | 1 | 2026-06-30 | View ranking |
| FrontierSWE | Unversioned | dominance | 1 | 1 | 2026-06-16 | View ranking |
| FrontierScience | Unversioned | score | 1 | 1 | Not documented | View ranking |
| GDP.PDF | Unversioned | all-pass rate | 1 | 1 | Not documented | View ranking |
| GDP.pdf | Unversioned | accuracy | 1 | 1 | 2026-08-13 | View ranking |
| GDPval-AA | Unversioned | win rate | 1 | 1 | 2026-06-09 | View ranking |
| GPQA | Unversioned | accuracy | 1 | 1 | 2026-06-04 | View ranking |
| HLE-Diamond | Unversioned | score | 1 | 1 | 2026-10-15 | View ranking |
| HLE-Verified | Unversioned | accuracy | 1 | 1 | Not documented | View ranking |
| HMMT | February 2026 | accuracy | 1 | 1 | 2026-06-16 | View ranking |
| HMMT | November 2025 | accuracy | 1 | 1 | 2026-06-16 | View ranking |
| HR-Bench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| Harvey Legal Agent Benchmark | Unversioned | all-pass rate | 1 | 1 | Not documented | View ranking |
| HealthBench Professional | Unversioned | score | 1 | 1 | 2026-07-24 | View ranking |
| HorizonMath | Unversioned | pass@12 | 1 | 1 | Not documented | View ranking |
| IFEval | Unversioned | accuracy | 1 | 1 | 2026-06-30 | View ranking |
| IMOAnswerBench | Unversioned | accuracy | 1 | 1 | 2026-06-16 | View ranking |
| IMOAnswerBench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| IntPhys2 | Unversioned | score | 1 | 1 | 2026-10-15 | View ranking |
| JFBench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| JobBench | Unversioned | success rate | 1 | 1 | 2026-07-09 | View ranking |
| Kimi Claw 24/7 Bench | Unversioned | score | 1 | 1 | 2026-06-12 | View ranking |
| Kimi Code Bench | v2 | score | 1 | 1 | 2026-06-12 | View ranking |
| LAB-Bench | 2 | macro-average | 1 | 1 | Not documented | View ranking |
| LMArena Text-to-Video Arena | Unversioned | Elo | 1 | 1 | 2026-06-30 | View ranking |
| LVBench | Unversioned | pass@1 | 1 | 1 | Not documented | View ranking |
| LVBench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| LiveCodeBench | v6 | pass@1 | 1 | 1 | 2026-06-04 | View ranking |
| LiveCodeBench v6 | Unversioned | score | 1 | 1 | Not documented | View ranking |
| MCP Atlas | Unversioned | pass rate | 1 | 1 | 2026-08-05 | View ranking |
| MCP Mark Verified | Unversioned | success rate | 1 | 1 | 2026-06-12 | View ranking |
| MLS Bench Lite | Unversioned | score | 1 | 1 | 2026-06-12 | View ranking |
| MMLU | Unversioned | accuracy | 1 | 1 | Not documented | View ranking |
| MMLU-Pro | Unversioned | accuracy | 1 | 1 | 2026-06-04 | View ranking |
| MMMU-Pro | Unversioned | score | 1 | 1 | Not documented | View ranking |
| MRCR | 2 | score | 1 | 1 | Not documented | View ranking |
| MathArena Apex | Unversioned | score | 1 | 1 | Not documented | View ranking |
| MiMo Coding Bench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| OSWorld | 2.0 | partial reward | 1 | 1 | 2026-09-22 | View ranking |
| OSWorld | 2.0 | strict success rate | 1 | 1 | 2026-09-01 | View ranking |
| OmniDocBench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| OmniDocBench | 1.5 | score | 1 | 1 | Not documented | View ranking |
| PHYBench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| PLawBench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| PRBench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| PaperBench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| ParseBench | Unversioned | mean | 1 | 1 | Not documented | View ranking |
| PinchBench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| PostTrainBench | Unversioned | score | 1 | 1 | 2026-06-16 | View ranking |
| PostTrainBench | Unversioned | weighted average | 1 | 1 | Not documented | View ranking |
| ProgramBench | Unversioned | almost solved | 1 | 1 | Not documented | View ranking |
| ProgramBench | Unversioned | almost@1 | 1 | 1 | Not documented | View ranking |
| QwenQoderBench | Unversioned | avg@5 | 1 | 1 | Not documented | View ranking |
| QwenReactBench | Unversioned | Elo | 1 | 1 | Not documented | View ranking |
| QwenSVGBench | Unversioned | Elo | 1 | 1 | Not documented | View ranking |
| SEC-Bench Pro | Unversioned | pass@1 | 1 | 1 | Not documented | View ranking |
| SRE-Bench | Unversioned | success rate | 1 | 1 | 2026-09-03 | View ranking |
| SWE Marathon | Unversioned | pass@1 | 1 | 1 | 2026-07-08 | View ranking |
| SWE Marathon | Unversioned | resolve rate | 1 | 1 | 2026-06-16 | View ranking |
| SWE Marathon | 1.1 | resolve rate | 1 | 1 | 2026-07-16 | View ranking |
| SWE-Bench Multilingual | Unversioned | accuracy | 1 | 1 | Not documented | View ranking |
| SWE-Bench Multimodal | Unversioned | resolve rate | 1 | 1 | 2026-07-24 | View ranking |
| SWE-Bench Pro | 1 | score | 1 | 1 | 2026-10-16 | View ranking |
| SWE-Bench Pro | 2 | score | 1 | 1 | 2026-10-16 | View ranking |
| SWE-Bench Verified | Unversioned | accuracy | 1 | 1 | Not documented | View ranking |
| SWE-Bench Verified | Unversioned | resolve rate | 1 | 1 | 2026-08-10 | View ranking |
| SWE-Bench Verified | Unversioned | score | 1 | 1 | 2026-10-16 | View ranking |
| SWE-MM | Unversioned | score | 1 | 1 | Not documented | View ranking |
| SWE-Marathon | 1.1 | score | 1 | 1 | Not documented | View ranking |
| SWE-bench Pro | Unversioned | resolved | 1 | 1 | Not documented | View ranking |
| ScreenSpot-Pro | Unversioned | accuracy | 1 | 1 | 2026-09-03 | View ranking |
| SimpleVQA | Unversioned | score | 1 | 1 | 2026-10-15 | View ranking |
| SkillsBench | Unversioned | average over 3 runs | 1 | 1 | Not documented | View ranking |
| SkillsBench | 1.1 | avg@3 | 1 | 1 | Not documented | View ranking |
| SpreadsheetBench | 1 | score | 1 | 1 | Not documented | View ranking |
| SuperChem | Unversioned | score | 1 | 1 | Not documented | View ranking |
| Terminal Bench 2.0 | Unversioned | score | 1 | 1 | Not documented | View ranking |
| Terminal-Bench | 2.0 | score | 1 | 1 | Not documented | View ranking |
| Terminal-Bench | 2.1 | avg@10 | 1 | 1 | Not documented | View ranking |
| Terminal-Bench | 3.0 | avg@3 | 1 | 1 | Not documented | View ranking |
| Terminal-Bench | 3.0 | pass@1 | 1 | 1 | Not documented | View ranking |
| Terminal-Bench | 4.0 | success rate | 1 | 1 | 2026-09-03 | View ranking |
| Terminal-Bench Science | 0.1 | success rate | 1 | 1 | 2026-09-03 | View ranking |
| Tool-Decathlon | Unversioned | score | 1 | 1 | 2026-06-16 | View ranking |
| Toolathlon Verified | Unversioned | pass@1 | 1 | 1 | 2026-08-03 | View ranking |
| Toolathlon-Verified | Unversioned | success rate | 1 | 1 | 2026-07-09 | View ranking |
| Translation | Unversioned | score | 1 | 1 | Not documented | View ranking |
| USAMO | 2026 | score | 1 | 1 | Not documented | View ranking |
| Video-MME | Unversioned | score | 1 | 1 | Not documented | View ranking |
| Video-MME | 2 | score | 1 | 1 | 2026-10-15 | View ranking |
| VideoHolmes | Unversioned | score | 1 | 1 | Not documented | View ranking |
| WebArena-Verified | Unversioned | score | 1 | 1 | Not documented | View ranking |
| WideSearch | Unversioned | F1 | 1 | 1 | 2026-08-03 | View ranking |
| WideSearch | Unversioned | average item F1 | 1 | 1 | Not documented | View ranking |
| WideSearch | Unversioned | score | 1 | 1 | Not documented | View ranking |
| WildClawBench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| WorkSpaceBench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| WorldVQA | Unversioned | score | 1 | 1 | 2026-10-15 | View ranking |
| ZeroBench | Unversioned | score | 1 | 1 | 2026-10-15 | View ranking |
| olmOCR-bench | Unversioned | score | 1 | 1 | Not documented | View ranking |
| τ³-Telecom | Unversioned | accuracy | 1 | 1 | 2026-05-22 | View ranking |