Atlas

Compare Models

Grok 4.7 vs Qwen4-Max

Qwen4-Max leads by 3.1 capability points (≈95% interval +0.7 to +5.4): a clear lead over Grok 4.7 · P(higher) ≈ 100%

Wins 37 of 53 shared benchmarks (16 lost)

Capability vs. relative cost
Grok 4.7Qwen4-MaxClosed weightsOpen weights

Benchmark wins16 – 37 · 53 shared

Comparing matching setups only (2 benchmarks excluded) · suites counted once
← Grok 4.7Qwen4-Max →
Data: Artificial Analysis, Surge AI, Vals AI and 5 othersatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Qwen4-Max minus Grok 4.7.

Terminal-Bench 4.0 (Artificial Analysis)Software › Software engineering25.848.5+22.7
Humanity's Last Exam DiamondKnowledge & reasoning › General knowledge23.644.9+21.3
Terminal-Bench 4.0Software › Software engineering28.848.5+19.7
Arena Image-to-WebDevSoftware › Frontend & design15881674+18.7
LMArena Code Arena WebDevSoftware › Frontend & design16391750+14.8
CritPt v2 (agentic, avg@5)Science › Physical sciences51.466.0+14.6
KradleBenchGames23.011.0−12.0
Terminal-Bench-Science 0.1Science › Multidisciplinary11.423.3+11.9
LMArena Agent Arena Net Improvement (Steering Burden)Tool & computer use › General agents1.57.3+11.3
LLM Debate BenchmarkLanguage & writing › Language & wordplay16011679+11.1
Artificial Analysis Engineering IndexScience › Engineering41.850.1+8.3
Artificial Analysis Intelligence Index v5.0.1Composite indexes35.542.9+7.4
AutomationBench-AA 1.0.6Tool & computer use › Customer service & operations65.672.6+7.0
AA-AnalystAgent (pass^5)Professional › General knowledge work52.546.3−6.2
LMArena Agent Arena Steering BurdenTool & computer use › General agents-4.38.0+6.2
Finance Agent v2.1Professional › Finance, accounting & tax60.666.7+6.1
Legal Research BenchProfessional › Legal47.141.3−5.8
Artificial Analysis Economics IndexComposite indexes45.751.3+5.6
Code Migration v1.1 (Vals Index v2.2 subset)Software › Software engineering51.857.3+5.4
LMArena Agent Arena Praise vs ComplaintTool & computer use › General agents2.113.0+5.4
GDP.pdf (November 2026 rescoring)Professional › General knowledge work48.843.6−5.2
LiveBench 2.0 (2026-11-06 release)Composite indexes67.973.1+5.2
DualEntry Accounting AI BenchmarkProfessional › Finance, accounting & tax88.183.2−4.9
HANDBOOK.md Agents v1.1Tool & computer use › Customer service & operations44.039.4−4.6
Vals Index v2.2Composite indexes58.663.2+4.6
GDPval-AA v2.1Professional › General knowledge work17171809+4.0
DAYJOB: TaxProfessional › Finance, accounting & tax5.08.8+3.8
sudo L7 (October 2026 rescoring)Software › Software engineering12.215.6+3.4
DAYJOB: Healthcare (October 2026 rescoring)Health & medicine27.624.4−3.2
Vals Finance IndexProfessional › Finance, accounting & tax58.862.0+3.2
GDP.pdf (Artificial Analysis, All-pass)Professional › General knowledge work23.226.2+3.0
HLAB-Vals-1.1Professional › Legal18.321.3+2.9
AA-Briefcase v1.1 EloProfessional › General knowledge work16341686+2.8
LMArena Agent Arena Confirmed SuccessTool & computer use › General agents3.58.8+2.7
LMArena Agent Arena Bash ReliabilityTool & computer use › General agents8.23.4−2.4
Artificial Analysis Finance & Accounting IndexProfessional › Finance, accounting & tax51.754.1+2.4
DAYJOB: AccountingProfessional › Finance, accounting & tax20.022.4+2.4
Artificial Analysis Legal IndexProfessional › Legal47.249.6+2.4
FrontierCode 2.0 (Artificial Analysis)Software › Software engineering28.726.5−2.2
Artificial Analysis Strategy & Ops IndexProfessional › General knowledge work54.856.9+2.0
Vals Legal IndexProfessional › Legal72.770.7−2.0
AA-Omniscience v1.1 IndexKnowledge & reasoning › General knowledge31.227.7−1.8
ComplexConstraints (November 2026 rescoring)Language & writing › Instruction following60.859.1−1.7
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & medicine21.119.4−1.7
Artificial Analysis Healthcare & Medical IndexHealth & medicine46.648.2+1.6
GDP.xlsxProfessional › General knowledge work22.323.7+1.4
DAYJOB: Finance (October 2026 rescoring)Professional › Finance, accounting & tax26.525.3−1.2
Tax Agent BenchProfessional › Finance, accounting & tax69.070.0+1.0
Chartography (November 2026 rescoring)Perception & media › Documents & charts74.675.3+0.7
Excel Modeling Benchmark v1.1Professional › Finance, accounting & tax71.570.8−0.7
EnterpriseBench: CoreCraftTool & computer use › Customer service & operations68.268.7+0.5
Vibe Code Bench v1.1Software › Frontend & design86.286.7+0.5
LMArena Agent Arena Tool HallucinationTool & computer use › Tool calling1.21.20.0
Data: Artificial Analysis, Surge AI, Vals AI and 5 othersatlas.village.dev
Win counts use each benchmark’s declared score direction. Represented suite slices and the derived Capability score are excluded from headline totals. Missing or incompatible setups are excluded, never counted as losses.
Loading Atlas data…