Atlas

Compare Models

GPT-6.2 Sol vs Qwen4-Max

GPT-6.2 Sol leads by 3.3 capability points (≈95% interval +0.6 to +5.9): a clear lead over Qwen4-Max · P(higher) ≈ 99%

Wins 38 of 53 shared benchmarks (15 lost)

Capability vs. relative cost
GPT-6.2 SolQwen4-MaxClosed weightsOpen weights

Benchmark wins38 – 15 · 53 shared

Comparing matching setups only (3 benchmarks excluded) · suites counted once
← GPT-6.2 SolQwen4-Max →
Data: Artificial Analysis, Surge AI, Vals AI and 6 othersatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Qwen4-Max minus GPT-6.2 Sol.

KradleBenchGames67.911.0−56.9
Terminal-Bench-Science 0.1Science › Multidisciplinary66.723.3−43.3
EnterpriseBench: CoreCraftTool & computer use › Customer service & operations36.468.7+32.3
Code Migration v1.1 (Vals Index v2.2 subset)Software › Software engineering85.357.3−28.1
EnterpriseOps-Gym-AATool & computer use › Customer service & operations26.348.5+22.2
sudo L7 (October 2026 rescoring)Software › Software engineering36.115.6−20.5
Humanity's Last Exam DiamondKnowledge & reasoning › General knowledge64.044.9−19.1
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & medicine37.219.4−17.8
FrontierCode 2.0 (Artificial Analysis)Software › Software engineering42.826.5−16.3
DAYJOB: Law (November 2026 release)Professional › Legal32.717.5−15.2
Artificial Analysis Intelligence Index v5.0.1Composite indexes57.342.9−14.4
Chartography (November 2026 rescoring)Perception & media › Documents & charts87.775.3−12.4
CritPt v2 (agentic, avg@5)Science › Physical sciences78.366.0−12.3
Terminal-Bench 4.0Software › Software engineering60.648.5−12.1
Artificial Analysis Engineering IndexScience › Engineering61.250.1−11.1
ComplexConstraints (November 2026 rescoring)Language & writing › Instruction following69.559.1−10.4
HLAB-Vals-1.1Professional › Legal12.121.3+9.2
LMArena Agent Arena Net Improvement (Steering Burden)Tool & computer use › General agents11.37.3−7.9
Finance Agent v2.1Professional › Finance, accounting & tax58.966.7+7.8
Terminal-Bench 4.0 (Artificial Analysis)Software › Software engineering56.148.5−7.6
AA-AnalystAgent (pass^5)Professional › General knowledge work53.846.3−7.5
FrontierCUA v1 pass@1Tool & computer use › Computer & browser use25.017.5−7.5
DAYJOB: Finance (October 2026 rescoring)Professional › Finance, accounting & tax32.825.3−7.5
LiveBench 2.0 (2026-11-06 release)Composite indexes80.373.1−7.2
DAYJOB: Healthcare (October 2026 rescoring)Health & medicine17.224.4+7.2
Tax Agent BenchProfessional › Finance, accounting & tax62.970.0+7.1
AA-Omniscience v1.1 IndexKnowledge & reasoning › General knowledge40.427.7−6.3
Artificial Analysis Economics IndexComposite indexes57.451.3−6.0
AA-Briefcase v1.1 EloProfessional › General knowledge work15801686+5.8
Artificial Analysis Legal IndexProfessional › Legal55.049.6−5.5
Vals Finance IndexProfessional › Finance, accounting & tax56.762.0+5.3
DAYJOB: AccountingProfessional › Finance, accounting & tax27.622.4−5.2
GDP.pdf (November 2026 rescoring)Professional › General knowledge work48.843.6−5.2
LMArena Agent Arena Praise vs ComplaintTool & computer use › General agents23.313.0−5.1
Artificial Analysis Healthcare & Medical IndexHealth & medicine53.348.2−5.1
Agents' Last Exam v1.1 (ALE v1.1)Professional › General knowledge work41.846.7+4.9
HANDBOOK.md Agents v1.1Tool & computer use › Customer service & operations34.539.4+4.9
Legal Research BenchProfessional › Legal37.041.3+4.3
Vals Legal IndexProfessional › Legal66.770.7+4.0
DAYJOB: TaxProfessional › Finance, accounting & tax4.88.8+4.0
GDP.pdf (Artificial Analysis, All-pass)Professional › General knowledge work30.026.2−3.8
Excel Modeling Benchmark v1.1Professional › Finance, accounting & tax74.370.8−3.5
Artificial Analysis Finance & Accounting IndexProfessional › Finance, accounting & tax57.254.1−3.1
Vals Index v2.2Composite indexes66.163.2−2.9
LMArena Agent Arena Steering BurdenTool & computer use › General agents13.48.0−2.7
Artificial Analysis Strategy & Ops IndexProfessional › General knowledge work59.456.9−2.6
GDP.xlsxProfessional › General knowledge work22.323.7+1.4
AutomationBench-AA 1.0.6Tool & computer use › Customer service & operations71.372.6+1.3
LMArena Agent Arena Confirmed SuccessTool & computer use › General agents10.98.8−1.0
DualEntry Accounting AI BenchmarkProfessional › Finance, accounting & tax84.283.2−1.0
Vibe Code Bench v1.1Software › Frontend & design87.686.7−1.0
LMArena Agent Arena Bash ReliabilityTool & computer use › General agents4.33.4−0.5
LMArena Agent Arena Tool HallucinationTool & computer use › Tool calling1.21.20.0
Data: Artificial Analysis, Surge AI, Vals AI and 6 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…