Evidence coverage

2666 published observations across 285 measures. Capability scoring groups compatible evidence into 43 benchmark families.

A published result contributes only when a matching reference-panel result is available. Missing evidence stays unknown. Coverage describes breadth, separately from performance.

Coverage across our labs

LabModels with resultsPublished measuresLaunch sources
OpenAI31 / 3720916
Anthropic11 / 1121717
xAI3 / 7181
Google21 / 341098
Meta11 / 14332
Mistral16 / 27151
Moonshot3 / 71197
Qwen28 / 81914
DeepSeek4 / 15282
NVIDIA2 / 10411
Z.ai17 / 52937
MiniMax6 / 13642
Cohere10 / 26181
Amazon5 / 6191
Microsoft9 / 24191
Xiaomi3 / 1130
StepFun2 / 420
LG AI Research5 / 1330
Upstage4 / 820
ByteDance1 / 1610
Tencent1 / 1320
Baidu1 / 2010
OpenBMB1 / 110
Agnes AI1 / 110

Models and contributing families

Raw published counts include all source types. Contributing and independent family counts follow the selected evidence policy.

451 models match.

ModelPublished measuresContributing familiesWith independent supportExplore
Grok-4.51352Published results
Muse Spark 1.21251Published results
Gemini 3.5 Flash642Published results
Gemini 3.5 Flash-Lite444Published results
GLM-5.3-Flash444Published results
GPT-4o Mini444Published results
MiniMax-M2.5444Published results
o3444Published results
Claude Opus 4.5333Published results
Command A Reasoning 08 20251030Published results
Command A Vision 07 2025430Published results
DeepSeek-R1-0528333Published results
DeepSeek-V3.2333Published results
GLM-4.5333Published results
GPT-4.1333Published results
GPT-5.2333Published results
gpt-oss-120b333Published results
MiniMax-M2333Published results
Mistral Medium 3.1732Published results
o3-mini333Published results
o4-mini333Published results
Qwen3-235B-A22B333Published results
Qwen3-235B-A22B-Instruct-2507333Published results
aya-expanse-8b222Published results
Devstral 2322Published results

Family coverage

Benchmark familyCapabilityModels contributingWith independent support
OSWorldAgentic196
Webarena VerifiedAgentic40
BrowsecompAgentic220
ToolathlonAgentic150
MCP-AtlasAgentic190
AutomationBenchAgentic190
DeepSearchQAAgentic100
GDPvalAgentic6041
Bfcl V4Agentic120
Apex AgentsAgentic100
OfficeQAAgentic110
Humanity’s Last ExamHard reasoning318
GPQAHard reasoning4823
ARC-AGIHard reasoning4848
AimeHard reasoning220
FrontierMathHard reasoning100
Imo AnswerbenchHard reasoning50
Terminal-Bench ScienceHard reasoning50
SWE-bench VerifiedCoding3819
SWE-bench ProCoding204
SWE-bench MultilingualCoding90
DeepSWECoding3128
LiveCodeBenchCoding259
Terminal-BenchCoding412
ScicodeCoding70
Human preference (Arena)Human pref115115
MMLU-ProKnowledge6852
GmmluKnowledge30
MiluKnowledge30
Simpleqa VerifiedKnowledge20
MmmuMultimodal20
Mmmu ProMultimodal240
CharXivMultimodal160
MathvistaMultimodal20
ChartographyMultimodal30
LvbenchMultimodal60
ScreenSpotMultimodal120
Ocrbench V2 Average AccuracyMultimodal120
MRCRLong context120
LongbenchLong context80
Ruler 1MLong context20
GraphwalksLong context70
Aa LcrLong context50
Model evidence matrix

Showing the same 51–75 models as the list above.

ModelOSWorldWebarena VerifiedBrowsecompToolathlonMCP-AtlasAutomationBenchDeepSearchQAGDPvalBfcl V4Apex AgentsOfficeQAHumanity’s Last ExamGPQAARC-AGIAimeFrontierMathImo AnswerbenchTerminal-Bench ScienceSWE-bench VerifiedSWE-bench ProSWE-bench MultilingualDeepSWELiveCodeBenchTerminal-BenchScicodeHuman preference (Arena)MMLU-ProGmmluMiluSimpleqa VerifiedMmmuMmmu ProCharXivMathvistaChartographyLvbenchScreenSpotOcrbench V2 Average AccuracyMRCRLongbenchRuler 1MGraphwalksAa Lcr
Grok-4.5UnknownUnknownUnknownUnknownUnknownUnknownUnknown0.0
Provider reports · 3/3 reference peers
Unknown0.0
Provider reports · 3/3 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown17.6
Independent support · 27/27 reference peers
Unknown0.0
Provider reports · 3/3 reference peers
Unknown87.7
Independent support · 114/114 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
Muse Spark 1.2UnknownUnknownUnknownUnknownUnknown0.0
Provider reports · 1/1 reference peers
0.0
Provider reports · 1/1 reference peers
0.0
Provider reports · 1/1 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown40.7
Independent support · 27/27 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown0.0
Provider reports · 1/1 reference peers
UnknownUnknownUnknownUnknown
Gemini 3.5 FlashUnknownUnknownUnknownUnknownUnknown16.7
Provider reports · 6/6 reference peers
Unknown14.3
Provider reports · 7/7 reference peers
UnknownUnknownUnknownUnknownUnknown73.4
Independent support · 47/47 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknown11.1
Independent support · 27/27 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
Gemini 3.5 Flash-LiteUnknownUnknownUnknownUnknownUnknownUnknownUnknown60.0
Independent support · 40/40 reference peers
UnknownUnknownUnknownUnknown9.1
Independent support · 22/22 reference peers
44.7
Independent support · 47/47 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown80.7
Independent support · 114/114 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
GLM-5.3-FlashUnknownUnknownUnknownUnknownUnknownUnknownUnknown97.5
Independent support · 40/40 reference peers
UnknownUnknownUnknownUnknown27.3
Independent support · 22/22 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown55.6
Independent support · 27/27 reference peers
UnknownUnknownUnknown91.2
Independent support · 114/114 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
GPT-4o MiniUnknownUnknownUnknownUnknownUnknownUnknownUnknown5.0
Independent support · 40/40 reference peers
UnknownUnknownUnknownUnknownUnknown4.3
Independent support · 47/47 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown0.0
Independent support · 8/8 reference peers
UnknownUnknownUnknown33.3
Independent support · 51/51 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
MiniMax-M2.5UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown31.9
Independent support · 47/47 reference peers
UnknownUnknownUnknownUnknown97.2
Independent support · 18/18 reference peers
UnknownUnknownUnknownUnknownUnknownUnknown47.4
Independent support · 114/114 reference peers
56.9
Independent support · 51/51 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
o30.0
Independent support · 5/5 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown40.4
Independent support · 47/47 reference peers
UnknownUnknownUnknownUnknown50.0
Independent support · 18/18 reference peers
UnknownUnknownUnknown87.5
Independent support · 8/8 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
Claude Opus 4.5UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown57.1
Independent support · 7/7 reference peers
UnknownUnknownUnknownUnknownUnknownUnknown83.3
Independent support · 18/18 reference peers
UnknownUnknownUnknownUnknownUnknownUnknown88.6
Independent support · 114/114 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
Command A Reasoning 08 2025UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown0.0
Provider reports · 1/1 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown0.0
Provider reports · 1/1 reference peers
0.0
Provider reports · 1/1 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
Command A Vision 07 2025UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown0.0
Provider reports · 1/1 reference peers
Unknown0.0
Provider reports · 1/1 reference peers
0.0
Provider reports · 1/1 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
DeepSeek-R1-0528UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown75.0
Independent support · 8/8 reference peers
UnknownUnknown61.4
Independent support · 114/114 reference peers
74.5
Independent support · 51/51 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
DeepSeek-V3.2UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown25.5
Independent support · 47/47 reference peers
UnknownUnknownUnknownUnknown66.7
Independent support · 18/18 reference peers
UnknownUnknownUnknownUnknownUnknownUnknown64.0
Independent support · 114/114 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
GLM-4.5UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown33.3
Independent support · 18/18 reference peers
UnknownUnknownUnknownUnknownUnknownUnknown56.1
Independent support · 114/114 reference peers
78.4
Independent support · 51/51 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
GPT-4.1UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown14.3
Independent support · 7/7 reference peers
Unknown10.6
Independent support · 47/47 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown64.7
Independent support · 51/51 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
GPT-5.2UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown55.3
Independent support · 47/47 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown69.3
Independent support · 114/114 reference peers
92.2
Independent support · 51/51 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
gpt-oss-120bUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown4.5
Independent support · 22/22 reference peers
UnknownUnknownUnknownUnknownUnknown5.6
Independent support · 18/18 reference peers
UnknownUnknownUnknownUnknownUnknownUnknown35.1
Independent support · 114/114 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
MiniMax-M2UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown55.6
Independent support · 18/18 reference peers
UnknownUnknownUnknownUnknownUnknownUnknown32.5
Independent support · 114/114 reference peers
66.7
Independent support · 51/51 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
Mistral Medium 3.1UnknownUnknown0.0
Provider reports · 2/2 reference peers
UnknownUnknownUnknownUnknown30.0
Independent support · 40/40 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown54.4
Independent support · 114/114 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
o3-miniUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown21.3
Independent support · 47/47 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown50.0
Independent support · 8/8 reference peers
UnknownUnknown34.2
Independent support · 114/114 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
o4-miniUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown38.3
Independent support · 47/47 reference peers
UnknownUnknownUnknownUnknown16.7
Independent support · 18/18 reference peers
UnknownUnknownUnknown100.0
Independent support · 8/8 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
Qwen3-235B-A22BUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown37.5
Independent support · 8/8 reference peers
UnknownUnknown42.1
Independent support · 114/114 reference peers
41.2
Independent support · 51/51 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
Qwen3-235B-A22B-Instruct-2507UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown13.8
Independent support · 47/47 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown62.3
Independent support · 114/114 reference peers
72.5
Independent support · 51/51 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
aya-expanse-8bUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown13.2
Independent support · 114/114 reference peers
5.9
Independent support · 51/51 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown
Devstral 2UnknownUnknownUnknownUnknownUnknownUnknownUnknown45.0
Independent support · 40/40 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown63.9
Independent support · 19/19 reference peers
UnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknownUnknown

Family definitions and scoring rules · Compare capabilities · Compare published evidence