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
Area← 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.
| KradleBench | Games | 43.9 | 11.0 | −32.9 |
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 49.5 | 23.3 | −26.2 |
| Code Migration v1.1 (Vals Index v2.2 subset) | Software › Software engineering | 82.7 | 57.3 | −25.4 |
| GDP.xlsx | Professional › General knowledge work | 40.9 | 23.7 | −17.2 |
| sudo L7 (October 2026 rescoring) | Software › Software engineering | 31.7 | 15.6 | −16.1 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 73.7 | 59.1 | −14.6 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 80.6 | 66.0 | −14.6 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 58.8 | 44.9 | −13.9 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 38.5 | 25.3 | −13.2 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 39.0 | 26.5 | −12.5 |
| Legal Research Bench | Professional › Legal | 53.8 | 41.3 | −12.5 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 13.5 | 7.3 | −12.2 |
| Chartography (November 2026 rescoring) | Perception & media › Documents & charts | 87.2 | 75.3 | −11.9 |
| DAYJOB: Accounting | Professional › Finance, accounting & tax | 33.6 | 22.4 | −11.2 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 54.2 | 43.6 | −10.6 |
| Artificial Analysis Engineering Index | Science › Engineering | 60.1 | 50.1 | −10.0 |
| HLAB-Vals-1.1 | Professional › Legal | 11.3 | 21.3 | +10.0 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 46.3 | 27.7 | −9.3 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 78.4 | 70.0 | −8.4 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 29.3 | 13.0 | −8.1 |
| DAYJOB: Law (November 2026 release) | Professional › Legal | 9.7 | 17.5 | +7.8 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 27.2 | 19.4 | −7.8 |
| Artificial Analysis Legal Index | Professional › Legal | 56.7 | 49.6 | −7.2 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 32.8 | 26.2 | −6.6 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1569 | 1686 | +6.4 |
| LMArena Code Arena WebDev | Software › Frontend & design | 1703 | 1750 | +6.3 |
| Terminal-Bench 4.0 | Software › Software engineering | 54.0 | 48.5 | −5.6 |
| Vals Index v2.2 | Composite indexes | 68.5 | 63.2 | −5.3 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 44.6 | 39.4 | −5.2 |
| DAYJOB: Tax | Professional › Finance, accounting & tax | 3.6 | 8.8 | +5.2 |
| Artificial Analysis Healthcare & Medical Index | Health & medicine | 53.3 | 48.2 | −5.1 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 53.5 | 48.5 | −5.1 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 75.8 | 70.8 | −5.0 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 17.4 | 8.0 | −4.7 |
| Vals Finance Index | Professional › Finance, accounting & tax | 66.1 | 62.0 | −4.1 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | -4.5 | 3.4 | +3.9 |
| Artificial Analysis Finance & Accounting Index | Professional › Finance, accounting & tax | 58.1 | 54.1 | −3.9 |
| LiveBench 2.0 (2026-11-06 release) | Composite indexes | 77.0 | 73.1 | −3.9 |
| Artificial Analysis Economics Index | Composite indexes | 55.0 | 51.3 | −3.7 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 15.7 | 8.8 | −3.4 |
| Vals Legal Index | Professional › Legal | 73.6 | 70.7 | −2.9 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 74.4 | 72.6 | −1.9 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 41.2 | 42.9 | +1.7 |
| Finance Agent v2.1 | Professional › Finance, accounting & tax | 65.1 | 66.7 | +1.6 |
| Vibe Code Bench v1.1 | Software › Frontend & design | 87.8 | 86.7 | −1.1 |
| Artificial Analysis Strategy & Ops Index | Professional › General knowledge work | 57.7 | 56.9 | −0.9 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 24.0 | 24.4 | +0.4 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 0.7 | 1.2 | +0.2 |
| DualEntry Accounting AI Benchmark | Professional › Finance, accounting & tax | 83.2 | 83.2 | 0.0 |
Data: Artificial Analysis, Surge AI, Vals AI and 4 othersatlas.village.dev