Evaluation sources
Verify results against original evaluation pages or published data. Listing a source does not mean its results have been synced or included in scoring.
| Original source | |||||
|---|---|---|---|---|---|
| AA Index | Artificial Analysis | Text | Coding | Cross-reference | artificialanalysis.ai |
| AA-Analyst Agent | Artificial Analysis | Text | Professional work | Participating | artificialanalysis.ai |
| AA-Briefcase | Artificial Analysis | Text | Professional work | Participating | artificialanalysis.ai |
| AA-LCR | Artificial Analysis | Text | Knowledge & language | Participating | artificialanalysis.ai |
| AA-Omniscience | Artificial Analysis | Text | Knowledge & language | Participating | artificialanalysis.ai |
| Agents' Last Exam | Agents’ Last Exam | Text | Professional work | Observing | agents-last-exam.org |
| Aider Polyglot | Aider | Text | Coding | Observing | aider.chat |
| APEX-Agents 1.1 | Mercor | Text | Professional work | Participating | www.mercor.com |
| ARC-AGI-1 | ARC Prize | Text | Reasoning | Observing | arcprize.org |
| ARC-AGI-2 | ARC Prize | Text | Reasoning | Observing | github.com |
| Arena Agent Overall | LMArena | Text | Professional work | Observing | huggingface.co |
| Arena Creative Preference | LMArena | Text | Writing & design | Reference only | lmarena.ai |
| Arena Text Overall | LMArena | Text | Human preference & composite | Participating | huggingface.co |
| Arena WebDev | LMArena | Text | Coding | Reference only | lmarena.ai |
| Artificial Analysis Coding Agent Index | Artificial Analysis | Text | Professional work | Observing | artificialanalysis.ai |
| Artificial Analysis Coding Index | Artificial Analysis | Text | Knowledge & language | Observing | artificialanalysis.ai |
| Artificial Analysis Intelligence Index | Artificial Analysis | Text | Knowledge & language | Observing | artificialanalysis.ai |
| AutomationBench | Zapier | Text | Professional work | Observing | github.com |
| AutomationBench-AA | Artificial Analysis | Text | Professional work | Participating | artificialanalysis.ai |
| BrowseComp | OpenAI | Text | Knowledge & language | Observing | openai.com |
| BullshitBench v2 | BullshitBench | Text | Knowledge & language | Participating | raw.githubusercontent.com |
| Chess Puzzles | Epoch AI | Text | Reasoning | Participating | epoch.ai |
| Claw-eval | Claw-Eval | Text | Knowledge & language | Observing | github.com |
| ClawEval | Claw-Eval | Text | Knowledge & language | Observing | github.com |
| Code Migration | Vals AI | Text | Coding | Participating | www.vals.ai |
| CritPt | Artificial Analysis | Text | Reasoning | Participating | artificialanalysis.ai |
| CTI-REALM | Microsoft | Text | Knowledge & language | Observing | github.com |
| CyberBench Patch v1.1 | Vals AI | Text | Coding | Observing | www.vals.ai |
| DeepSWE | Datacurve | Text | Coding | Observing | github.com |
| Finance Agent | Vals AI | Text | Professional work | Participating | www.vals.ai |
| Finance Agent | Vals AI | Text | Professional work | Observing | github.com |
| FinanceAgent | Vals AI | Text | Professional work | Observing | github.com |
| FORTE | AGI-Eval | Text | Knowledge & language | Observing | github.com |
| FrontierCode | Cognition | Text | Coding | Observing | cognition.com |
| FrontierMath | Epoch AI | Text | Reasoning | Observing | epoch.ai |
| FrontierMath v2 Tiers 1–3 | Epoch AI | Text | Reasoning | Participating | epoch.ai |
| FrontierSWE | FrontierSWE | Text | Coding | Observing | www.frontierswe.com |
| GDPval | OpenAI | Text | Professional work | Observing | openai.com |
| GDPval-AA | Artificial Analysis | Text | Professional work | Participating | artificialanalysis.ai |
| GDPval-AA | Artificial Analysis | Text | Professional work | Observing | artificialanalysis.ai |
| GeneBench | OpenAI | Text | Knowledge & language | Observing | cdn.openai.com |
| GPQA | GPQA authors | Text | Reasoning | Observing | github.com |
| GPQA Diamond | Scale | Text | Reasoning | Observing | github.com |
| Harvey Legal Agent Benchmark | Vals AI / Harvey | Text | Professional work | Participating | www.vals.ai |
| Humanity's Last Exam | Center for AI Safety / Scale AI | Text | Knowledge & language | Observing | github.com |
| Humanity’s Last Exam | Center for AI Safety / Scale AI | Text | Knowledge & language | Observing | github.com |
| Humanity’s Last Exam | Artificial Analysis | Text | Reasoning | Participating | artificialanalysis.ai |
| IFBench | Ai2 | Text | Knowledge & language | Observing | github.com |
| IFBench（Artificial Analysis） | Artificial Analysis | Text | Knowledge & language | Observing | artificialanalysis.ai |
| IFEval | Google Research | Text | Knowledge & language | Observing | github.com |
| JobBench | JobBench | Text | Professional work | Observing | job-bench.github.io |
| Kimi Claw 24/7 Bench | Moonshot AI | Text | Knowledge & language | Observing | huggingface.co |
| Kimi Code Bench | Moonshot AI | Text | Coding | Observing | huggingface.co |
| LiveBench Coding | LiveBench | Text | Coding | Observing | livebench.ai |
| LiveBench Data Analysis | LiveBench | Text | Professional work | Observing | livebench.ai |
| LiveBench General | LiveBench | Text | Reasoning | Cross-reference | livebench.ai |
| LiveBench Language & Instructions | LiveBench | Text | Knowledge & language | Observing | livebench.ai |
| LiveBench Reasoning & Mathematics | LiveBench | Text | Reasoning | Observing | livebench.ai |
| LiveCodeBench | LiveCodeBench | Text | Coding | Observing | livecodebench.github.io |
| LiveCodeBench Pro | LiveCodeBench Pro | Text | Coding | Observing | github.com |
| Long Context Reasoning | Artificial Analysis | Text | Reasoning | Observing | artificialanalysis.ai |
| MCP Atlas | Scale AI | Text | Professional work | Observing | labs.scale.com |
| MCP Mark Verified | MCPMark / Moonshot AI | Text | Professional work | Observing | huggingface.co |
| MedCode | Vals AI | Text | Professional work | Observing | www.vals.ai |
| MLCR-AA（Artificial Analysis） | Artificial Analysis | Text | Knowledge & language | Participating | artificialanalysis.ai |
| MLS Bench Lite | MLS-Bench | Text | Knowledge & language | Observing | github.com |
| MMLU-Pro | TIGER AI Lab | Text | Knowledge & language | Observing | github.com |
| MRCRv2 | OpenAI | Text | Knowledge & language | Observing | huggingface.co |
| Mystery Game Puzzles | Epoch AI | Text | Reasoning | Participating | epoch.ai |
| MysteryMechanism | Vals AI | Text | Reasoning | Participating | www.vals.ai |
| NL2Repo | MAP | Text | Coding | Observing | github.com |
| OSWorld | XLANG Lab | Text | Professional work | Observing | github.com |
| OSWorld-Verified | XLANG Lab | Text | Professional work | Observing | github.com |
| Program Bench | Meta | Text | Knowledge & language | Observing | github.com |
| ProgramBench | Vals AI | Text | Coding | Participating | www.vals.ai |
| SciCode | SciCode | Text | Coding | Observing | scicode-bench.github.io |
| SciCode | Artificial Analysis | Text | Coding | Participating | artificialanalysis.ai |
| SimpleQA Verified | Epoch AI | Text | Knowledge & language | Participating | epoch.ai |
| SWE Bench Pro | Scale AI | Text | Coding | Observing | github.com |
| SWE-Atlas Codebase QnA | Scale AI | Text | Coding | Observing | labs.scale.com |
| SWE-Atlas Refactoring | Scale AI | Text | Coding | Observing | labs.scale.com |
| SWE-Atlas Test Writing | Scale AI | Text | Coding | Observing | labs.scale.com |
| SWE-Bench Multilingual | SWE-bench | Text | Chinese & multilingual | Observing | www.swebench.com |
| SWE-Bench Pro | Scale AI | Text | Coding | Observing | github.com |
| SWE-Bench Verified | SWE-bench | Text | Coding | Observing | www.swebench.com |
| Terminal Bench 2.0 | Terminal-Bench | Text | Coding | Observing | www.tbench.ai |
| Terminal Bench 2.1 | Terminal-Bench | Text | Coding | Observing | github.com |
| Terminal-Bench | Terminal-Bench | Text | Coding | Observing | www.tbench.ai |
| Terminal-Bench 2.1 | Terminal-Bench | Text | Coding | Observing | github.com |
| Terminal-Bench 4.0 (AA rerun) | Artificial Analysis | Text | Coding | Participating | artificialanalysis.ai |
| Terminal-Bench 4.0 (Vals rerun) | Vals AI | Text | Coding | Participating | www.vals.ai |
| Terminal-Bench Hard | Artificial Analysis | Text | Coding | Observing | artificialanalysis.ai |
| Terminal-Bench Science · AA | Artificial Analysis | Text | Reasoning | Participating | artificialanalysis.ai |
| Terminal-Bench Science（Vals 复跑） | Vals AI | Text | Reasoning | Participating | www.vals.ai |
| Toolathlon | HKUST NLP | Text | Professional work | Observing | github.com |
| Toolathlon-Verified | HKUST NLP / Moonshot AI | Text | Professional work | Observing | github.com |
| Vibe Code Bench 1–100 | Vals AI | Text | Coding | Observing | www.vals.ai |
| Vibe Code Bench v1.1 | Vals AI | Text | Coding | Observing | www.vals.ai |
| τ²-Bench Telecom | Sierra | Text | Knowledge & language | Observing | github.com |
| τ3 Banking | Sierra | Text | Knowledge & language | Observing | github.com |
| τ³-Banking | Mercor | Text | Professional work | Participating | www.mercor.com |