Atlas

Compare Models

Gemini 4 Hydrogen vs Qwen4-Max

Gemini 4 Hydrogen leads by 2.0 capability points (≈95% interval −0.6 to +4.6): statistically tied · P(higher) ≈ 94%

Wins 38 of 49 shared benchmarks (10 lost, 1 tied)

Capability vs. relative cost
Gemini 4 HydrogenQwen4-MaxClosed weightsOpen weights

Benchmark wins38 – 10 · 49 shared, 1 tied

Comparing matching setups only (2 benchmarks excluded) · suites counted once
← Gemini 4 HydrogenQwen4-Max →
Data: Artificial Analysis, Surge AI, Vals AI and 4 othersatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Qwen4-Max minus Gemini 4 Hydrogen.

KradleBenchGames43.911.0−32.9
Terminal-Bench-Science 0.1Science › Multidisciplinary49.523.3−26.2
Code Migration v1.1 (Vals Index v2.2 subset)Software › Software engineering82.757.3−25.4
GDP.xlsxProfessional › General knowledge work40.923.7−17.2
sudo L7 (October 2026 rescoring)Software › Software engineering31.715.6−16.1
ComplexConstraints (November 2026 rescoring)Language & writing › Instruction following73.759.1−14.6
CritPt v2 (agentic, avg@5)Science › Physical sciences80.666.0−14.6
Humanity's Last Exam DiamondKnowledge & reasoning › General knowledge58.844.9−13.9
DAYJOB: Finance (October 2026 rescoring)Professional › Finance, accounting & tax38.525.3−13.2
FrontierCode 2.0 (Artificial Analysis)Software › Software engineering39.026.5−12.5
Legal Research BenchProfessional › Legal53.841.3−12.5
LMArena Agent Arena Net Improvement (Steering Burden)Tool & computer use › General agents13.57.3−12.2
Chartography (November 2026 rescoring)Perception & media › Documents & charts87.275.3−11.9
DAYJOB: AccountingProfessional › Finance, accounting & tax33.622.4−11.2
GDP.pdf (November 2026 rescoring)Professional › General knowledge work54.243.6−10.6
Artificial Analysis Engineering IndexScience › Engineering60.150.1−10.0
HLAB-Vals-1.1Professional › Legal11.321.3+10.0
AA-Omniscience v1.1 IndexKnowledge & reasoning › General knowledge46.327.7−9.3
Tax Agent BenchProfessional › Finance, accounting & tax78.470.0−8.4
LMArena Agent Arena Praise vs ComplaintTool & computer use › General agents29.313.0−8.1
DAYJOB: Law (November 2026 release)Professional › Legal9.717.5+7.8
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & medicine27.219.4−7.8
Artificial Analysis Legal IndexProfessional › Legal56.749.6−7.2
GDP.pdf (Artificial Analysis, All-pass)Professional › General knowledge work32.826.2−6.6
AA-Briefcase v1.1 EloProfessional › General knowledge work15691686+6.4
LMArena Code Arena WebDevSoftware › Frontend & design17031750+6.3
Terminal-Bench 4.0Software › Software engineering54.048.5−5.6
Vals Index v2.2Composite indexes68.563.2−5.3
HANDBOOK.md Agents v1.1Tool & computer use › Customer service & operations44.639.4−5.2
DAYJOB: TaxProfessional › Finance, accounting & tax3.68.8+5.2
Artificial Analysis Healthcare & Medical IndexHealth & medicine53.348.2−5.1
Terminal-Bench 4.0 (Artificial Analysis)Software › Software engineering53.548.5−5.1
Excel Modeling Benchmark v1.1Professional › Finance, accounting & tax75.870.8−5.0
LMArena Agent Arena Steering BurdenTool & computer use › General agents17.48.0−4.7
Vals Finance IndexProfessional › Finance, accounting & tax66.162.0−4.1
LMArena Agent Arena Bash ReliabilityTool & computer use › General agents-4.53.4+3.9
Artificial Analysis Finance & Accounting IndexProfessional › Finance, accounting & tax58.154.1−3.9
LiveBench 2.0 (2026-11-06 release)Composite indexes77.073.1−3.9
Artificial Analysis Economics IndexComposite indexes55.051.3−3.7
LMArena Agent Arena Confirmed SuccessTool & computer use › General agents15.78.8−3.4
Vals Legal IndexProfessional › Legal73.670.7−2.9
AutomationBench-AA 1.0.6Tool & computer use › Customer service & operations74.472.6−1.9
Artificial Analysis Intelligence Index v5.0.1Composite indexes41.242.9+1.7
Finance Agent v2.1Professional › Finance, accounting & tax65.166.7+1.6
Vibe Code Bench v1.1Software › Frontend & design87.886.7−1.1
Artificial Analysis Strategy & Ops IndexProfessional › General knowledge work57.756.9−0.9
DAYJOB: Healthcare (October 2026 rescoring)Health & medicine24.024.4+0.4
LMArena Agent Arena Tool HallucinationTool & computer use › Tool calling0.71.2+0.2
DualEntry Accounting AI BenchmarkProfessional › Finance, accounting & tax83.283.20.0
Data: Artificial Analysis, Surge AI, Vals AI and 4 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…