Search AI, LLM, agent, multimodal, and AI-for-science benchmarks
Today's radar
Benchmarks with this name
Daily briefing
Questions for today
| Benchmark | Seed 2.1 ProByteDance · announced Jun 24, 2026 · 94 benchmarks | Seed 2.1 TurboByteDance · announced Jun 24, 2026 · 90 benchmarks | Qwen3.6 PlusQwen · announced Apr 2, 2026 · 81 benchmarks | Qwen3.5-122B-A10BQwen · announced Feb 24, 2026 · 81 benchmarks | Qwen3.5-27BQwen · announced Feb 24, 2026 · 81 benchmarks | Qwen3.5-35B-A3BQwen · announced Feb 24, 2026 · 81 benchmarks |
|---|---|---|---|---|---|---|
| Shared-benchmark averageover the 9 benchmarks all 6 models share, 0–100 | 69.2#1 | 67.4#2 | 65.4#3 | 65.4#3 | 65.2#5 | 64.6#6 |
| Reasoning & Knowledge 17 | ||||||
| AA-LCRArtificial Analysis | — | — | — | — | — | |
| AA-LCRLLM Stats | — | — | ||||
| AA-Omniscience AccuracyArtificial Analysis | — | — | — | — | — | |
| AA-Omniscience Non-Hallucination RateArtificial Analysis | — | — | — | — | — | |
| ArcAGI2LLM Stats | — | — | — | — | ||
| BrowseComp-zhLLM Stats | — | — | — | |||
| C-EvalLLM Stats | — | — | ||||
| Humanity’s Last ExamArtificial Analysis | — | — | — | — | — | |
| IncludeLLM Stats | — | — | ||||
| KINALLM Stats | — | — | — | — | ||
| LongBench v2LLM Stats | — | — | ||||
| MAXIFELLM Stats | — | — | ||||
| MSQALLM Stats | — | — | — | — | ||
| Multi-ChallengeLLM Stats | — | — | — | |||
| NOVA-63LLM Stats | — | — | ||||
| Seal-0LLM Stats | — | — | — | |||
| WMT24++LLM Stats | — | — | ||||
| Mathematics 29 | ||||||
| AIME 2026LLM Stats | — | — | — | — | — | |
| Beyond AIMELLM Stats | — | — | — | — | ||
| CodeForcesLLM Stats | — | — | — | |||
| DynaMathLLM Stats | ||||||
| EMMALLM Stats | — | — | — | — | ||
| FrontierScience OlympiadLLM Stats | — | — | — | — | ||
| HMMT 2025LLM Stats | — | — | ||||
| HMMT Feb 26LLM Stats | — | — | — | — | — | |
| HMMT25LLM Stats | — | — | ||||
| HorizonMathLLM Stats | — | — | — | — | ||
| Humanity's Last ExamLLM Stats | ||||||
| IMO 2025LLM Stats | — | — | — | — | — | |
| IMO-AnswerBenchLLM Stats | — | — | — | — | — | |
| IMOProof-AdvLLM Stats | — | — | — | — | — | |
| LiveBenchLLM Stats | — | — | — | — | — | |
| LiveMathematicianBenchLLM Stats | — | — | — | — | ||
| MathArena ApexLLM Stats | — | — | — | — | ||
| MathVerseLLM Stats | — | — | — | — | ||
| MathVisionLLM Stats | ||||||
| MathVistaLLM Stats | — | — | — | — | ||
| MathVista-MiniLLM Stats | — | — | — | |||
| MMLU-ProLLM Stats | — | — | ||||
| MMLU-ProXLLM Stats | — | — | ||||
| MMLU-ReduxLLM Stats | — | — | ||||
| MMMLULLM Stats | — | — | ||||
| PolyMATHLLM Stats | — | — | ||||
| SciCodeLLM Stats | — | — | — | — | ||
| SuperGPQALLM Stats | ||||||
| We-MathLLM Stats | — | — | — | — | — | |
| Coding 33 | ||||||
| AetherCodeLLM Stats | — | — | — | — | ||
| Artifacts BenchLLM Stats | — | — | — | — | ||
| Claw-EvalLLM Stats | — | — | — | — | — | |
| CyberGymLLM Stats | — | — | — | — | ||
| DeepSWELLM Stats | — | — | — | — | ||
| FrontierCSLLM Stats | — | — | — | — | ||
| FrontierSWELLM Stats | — | — | — | — | — | |
| FullStackBench enLLM Stats | — | — | — | |||
| FullStackBench zhLLM Stats | — | — | — | |||
| Image2FloorPlanLLM Stats | — | — | — | — | ||
| LiveCodeBench v6LLM Stats | — | — | ||||
| MCP AtlasLLM Stats | — | — | — | |||
| MVBenchLLM Stats | — | — | — | |||
| NL2RepoLLM Stats | — | — | — | |||
| OJBenchLLM Stats | — | — | — | |||
| PostTrainBenchLLM Stats | — | — | — | — | ||
| Program BenchLLM Stats | — | — | — | — | ||
| Repo EnvLLM Stats | — | — | — | — | ||
| SciCodeArtificial Analysis | — | — | — | — | — | |
| SeedClawBenchLLM Stats | — | — | — | — | ||
| SkillsBenchLLM Stats | — | — | — | — | — | |
| SWE-AtlasLLM Stats | — | — | — | — | ||
| SWE-bench MultilingualLLM Stats | — | — | — | — | — | |
| SWE-Bench ProLLM Stats | — | — | — | |||
| SWE-Bench VerifiedLLM Stats | — | — | ||||
| Terminal-Bench 2.0LLM Stats | — | — | ||||
| Terminal-Bench 2.1LLM Stats | — | — | — | — | ||
| Terminal-Bench HardArtificial Analysis | — | — | — | — | — | |
| Terminal-Bench v2.1Artificial Analysis | — | — | — | — | — | |
| Trae Code GenLLM Stats | — | — | — | — | ||
| Trae Error FixLLM Stats | — | — | — | — | ||
| Web BenchLLM Stats | — | — | — | — | ||
| WildClawBenchLLM Stats | — | — | — | — | ||
| Agents & Tools 35 | ||||||
| Agent Startup BenchLLM Stats | — | — | — | — | ||
| Agents' Last ExamLLM Stats | — | — | — | — | — | |
| AndroidWorld_SRLLM Stats | — | — | — | |||
| APEX-AgentsLLM Stats | — | — | — | — | ||
| AutomationBench-AAArtificial Analysis | — | — | — | — | — | |
| BFCL-V4LLM Stats | — | — | — | |||
| BrowseCompLLM Stats | — | |||||
| ClawEval-MMLLM Stats | — | — | — | — | ||
| CreativeWorkLLM Stats | — | — | — | — | ||
| DeepPlanningLLM Stats | — | — | ||||
| Doubao Multi-Turn BenchLLM Stats | — | — | — | — | ||
| Finance Agent v1.1LLM Stats | — | — | — | — | ||
| Finance Agent v2LLM Stats | — | — | — | — | — | |
| GameWorldLLM Stats | — | — | — | — | ||
| GDPvalLLM Stats | — | — | — | — | ||
| GDPval-AA v2 (Elo)Artificial Analysis | — | — | — | — | — | |
| GDPval-AA v2 (normalized score)Artificial Analysis | — | — | — | — | — | |
| MCP-MarkLLM Stats | — | — | — | — | — | |
| MobileWorldLLM Stats | — | — | — | — | ||
| OfficeQA ProLLM Stats | — | — | — | — | ||
| OneMillion BenchLLM Stats | — | — | — | — | ||
| OSWorldLLM Stats | — | — | — | — | ||
| OSWorld-VerifiedLLM Stats | — | — | ||||
| PresentBenchLLM Stats | — | — | — | — | ||
| ScreenSpot ProLLM Stats | — | — | ||||
| t2-benchLLM Stats | — | — | — | |||
| TAU3-BenchLLM Stats | — | — | — | — | — | |
| TIR-BenchLLM Stats | — | — | ||||
| ToolathlonLLM Stats | — | — | — | |||
| VITA-BenchLLM Stats | — | — | ||||
| WideSearchLLM Stats | — | — | ||||
| Workspace BenchLLM Stats | — | — | — | — | ||
| xDailyBenchLLM Stats | — | — | — | — | ||
| τ²-Bench TelecomArtificial Analysis | — | — | — | — | — | |
| τ³-BankingArtificial Analysis | — | — | — | — | — | |
| Science 9 | ||||||
| CharXiv-DLLM Stats | — | — | — | — | ||
| CharXiv-RLLM Stats | ||||||
| CritPtArtificial Analysis | — | — | — | — | — | |
| FrontierScience ResearchLLM Stats | — | — | — | — | ||
| Global PIQALLM Stats | — | — | ||||
| GPQALLM Stats | — | — | ||||
| GPQA DiamondArtificial Analysis | — | — | — | — | — | |
| IPhO 2025LLM Stats | — | — | — | — | — | |
| SuperChemLLM Stats | — | — | — | — | ||
| Multimodal 63 | ||||||
| AI2DLLM Stats | — | — | ||||
| BabyVisionLLM Stats | — | |||||
| BLINKLLM Stats | — | — | — | — | ||
| CC-OCRLLM Stats | — | — | ||||
| ChartQAProLLM Stats | — | — | — | — | ||
| ContPhyLLM Stats | — | — | — | — | ||
| CountBenchLLM Stats | — | — | ||||
| CrossVidLLM Stats | — | — | — | — | ||
| DUDELLM Stats | — | — | — | — | ||
| EmbSpatialBenchLLM Stats | — | |||||
| ERQALLM Stats | ||||||
| Hallusion BenchLLM Stats | — | — | — | |||
| HypersimLLM Stats | — | — | — | |||
| LingoQALLM Stats | — | — | — | |||
| LiveSports-3KLLM Stats | — | — | — | — | ||
| LongVideoBenchLLM Stats | — | — | — | — | ||
| LVBenchLLM Stats | — | |||||
| MeasureBenchLLM Stats | — | — | — | — | ||
| MedXpertQALLM Stats | — | — | — | |||
| MinervaLLM Stats | — | — | — | — | ||
| MLVULLM Stats | — | — | ||||
| MMBench-V1.1LLM Stats | — | — | — | |||
| MMLongBench-128KLLM Stats | — | — | — | — | ||
| MMLongBench-DocLLM Stats | — | — | ||||
| MMMULLM Stats | — | — | ||||
| MMMU-ProArtificial Analysis | — | — | — | — | — | |
| MMMU-ProLLM Stats | ||||||
| MMSIBenchLLM Stats | — | — | — | — | ||
| MMStarLLM Stats | — | — | ||||
| MMVULLM Stats | — | — | — | |||
| MotionBenchLLM Stats | — | — | — | — | ||
| NusceneLLM Stats | — | — | — | |||
| OCRBenchLLM Stats | — | — | — | |||
| OCRBench_V2LLM Stats | — | — | — | — | ||
| ODinWLLM Stats | — | — | ||||
| OmniDocBench 1.5LLM Stats | — | — | ||||
| OVBenchLLM Stats | — | — | — | — | ||
| OVOBenchLLM Stats | — | — | — | — | ||
| PMC-VQALLM Stats | — | — | — | |||
| RealWorldQALLM Stats | ||||||
| RefCOCO-avgLLM Stats | — | — | ||||
| RefSpatialBenchLLM Stats | — | — | — | |||
| SimpleVQALLM Stats | ||||||
| SlakeVQALLM Stats | — | — | — | |||
| SUNRGBDLLM Stats | — | — | — | |||
| TOMATOLLM Stats | — | — | — | — | ||
| TreeBenchLLM Stats | — | — | — | — | ||
| TVBenchLLM Stats | — | — | — | — | ||
| V*LLM Stats | — | — | ||||
| Video-MMELLM Stats | — | — | — | |||
| VideoHolmesLLM Stats | — | — | — | — | ||
| VideoMME w sub.LLM Stats | — | — | — | |||
| VideoMME w/o sub.LLM Stats | — | — | — | |||
| VideoMMMULLM Stats | — | — | ||||
| VideoSimpleQALLM Stats | — | — | — | — | ||
| VisFactorLLM Stats | — | — | — | — | ||
| VisuLogicLLM Stats | — | — | — | — | ||
| VLMsAreBiasedLLM Stats | — | — | — | — | ||
| VLMsAreBlindLLM Stats | — | — | — | |||
| WorldBenchLLM Stats | — | — | — | — | ||
| WorldVQALLM Stats | — | — | — | — | ||
| ZEROBenchLLM Stats | — | |||||
| ZEROBench-SubLLM Stats | — | — | — | |||
| Other 3 | ||||||
| IFBenchArtificial Analysis | — | — | — | — | — | |
| IFBenchLLM Stats | — | — | ||||
| IFEvalLLM Stats | — | — | ||||
2,950 benchmark records have no reported model score. Browse them in the Library
Leaderboard
Latest releases · 30 days
Recently released benchmarks, ranked by the attention they are receiving now: GitHub stars, Hugging Face paper upvotes and Hugging Face dataset downloads over the last 30 days.
Benchmark Frontier
Which difficult benchmarks have been tested most
Each mark is one benchmark record. Height counts distinct models with numeric scores. Source model IDs preserve evaluated configurations; repeated score rows add no models. Source documents counts distinct cited reports or registry pages through the same evidence collection. Both counts appear on hover.
The side view projects reported scores and the selected counts onto the left wall, leaving out time. The gold staircase compares only benchmarks with verified score scales. Within the dated 2024+ cohort, a benchmark is Pareto when no other eligible benchmark has an equal or lower normalized score and an equal or greater selected count, with at least one strict advantage. Dates select the cohort; they do not determine dominance. Scores and counts are never multiplied. The score slice cannot promote a dominated benchmark onto the frontier.
Height uses log1p(count); ticks, tooltips and Pareto use raw counts. The axis covers the full scored 2024+ cohort and stays fixed while filtering. Only declared percentage metrics with a known selected count enter Pareto; lower-is-better percentages become 100 minus the original value. Hollow points flag an unverified score scale or an unknown count, which is never treated as zero. Overlapping caps are spaced apart; hover or focus traces their exact coordinates. Arrow keys move through benchmarks. A shared scale does not establish equivalent test protocols or prove a benchmark is solved.
The timeline starts on January 1, 2024, inclusive. Dates use the benchmark's release first, then its earliest numeric LLM score report. When a source dates score entries by model release, the earliest entry supplies a clearly labelled model-release proxy; this is not a verified score-publication date. Crawl timestamps and adoption-only mentions never supply dates. Known dates before 2024 are excluded. Records with no date remain individually visible in a labelled area.
No scored benchmarks match these filters.
Most documented benchmarks
What the evidence shows Stated findings
Benchmarks by source documents
Each source document counts once per benchmark record. Repeated scores or mentions in the same document add no citations. This table covers all sources.
Audit the counts Source documents in the catalog
Source documents from every registry. Expand a document to inspect the benchmarks it records and open the original evidence.
Big picture
What we found
See what shows up most often and where it came from. Open the connections view when you want to look at a specific item.
Want more detail? See how everything connects
There are a lot of dots here. Pick one to see what it connects to, or open the matching results.
Saturation
Scores over time
Benchmark reported scores over time
How to read this chart
Every value that could be read verbatim from a cited document, placed at the date that document was published rather than at any evaluation date. The line directly links actual observations that set a new reported record among the values shown; it neither holds a score between reports nor extends past the final record. Test versions and run conditions can differ, so this is a reported-record path, not a like-for-like trend. When no newer number could be read, the gap is marked rather than drawn through. Whether a benchmark has saturated stays a reading you make, not a score this panel prints.
Trend
New by domain
Daily evidence and attention volume
Category tags overlap. Each bar is an independent count, not a part of a stacked total.
Releases only excludes records re-announced as an update to something already surfaced.
Dev checker
Source mix counts ranked evidence after scoring. Fetch health counts raw records returned before scoring, so a source can be ok and still empty.
| Date | Coverage (UTC) | Evidence | Source mix | Categories | Events | Attention | Fetch health |
|---|
Dashboard unavailable
The validated data file could not be loaded.
Try refreshing, or inspect the latest daily Issue while the dashboard rebuilds.
Open daily Issues ↗