Gemini 4 Hydrogen vs Muse Wonder
TryTop 2 now
Gemini 4 Hydrogen leads by 2.3 capability points (≈95% interval −0.2 to +4.7): statistically tied · P(higher) ≈ 96%
Wins 56 of 76 shared benchmarks (18 lost, 2 tied)
Capability vs. relative cost
Gemini 4 HydrogenMuse WonderClosed weightsOpen weights
Benchmark wins56 – 18 · 76 shared, 2 tied
Comparing matching setups only · suites counted once
Area← Gemini 4 HydrogenMuse Wonder →
Data: Artificial Analysis, Vals AI, Surge AI and 18 othersatlas.village.dev
Margins
Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Muse Wonder minus Gemini 4 Hydrogen.
| EBR-bench v4 (card ban, single agent) | Games | 70.5 | 31.4 | −39.0 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 13.5 | 0.6 | −25.2 |
| HiL-Bench (Human-in-Loop Benchmark) | Tool & computer use › General agents | 55.3 | 79.7 | +24.3 |
| SWE-Marathon v1.2 | Software › Software engineering | 58.1 | 38.1 | −20.0 |
| TaxCalcBench TY25 | Professional › Finance, accounting & tax | 44.0 | 24.0 | −20.0 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 58.8 | 39.0 | −19.8 |
| Humanity's Sixth Sense | Perception & media › Spatial & visual reasoning | 63.4 | 43.7 | −19.7 |
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 41.4 | 21.9 | −19.5 |
| GDP.xlsx | Professional › General knowledge work | 40.9 | 21.8 | −19.1 |
| FrontierMath Tier 4 (v2) - Private | Math | 87.8 | 70.7 | −17.1 |
| KradleBench | Games | 43.9 | 27.6 | −16.3 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 29.3 | -3.0 | −16.1 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 54.0 | 38.4 | −15.7 |
| sudo L7 (October 2026 rescoring) | Software › Software engineering | 31.7 | 16.1 | −15.6 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 44.6 | 29.5 | −15.1 |
| FrontierCode 2.0 - Pass Rate | Software › Software engineering | 43.2 | 28.8 | −14.4 |
| Mystery Game Puzzles | Games | 65.0 | 51.0 | −14.0 |
| FrontierCode 2.0 - Score | Software › Software engineering | 38.9 | 25.3 | −13.5 |
| Ramp SWE-Bench v2 | Software › Software engineering | 71.6 | 58.6 | −13.0 |
| Code Migration v1.1 (Vals Index v2.2 subset) | Software › Software engineering | 82.7 | 69.8 | −12.9 |
| HLAB-Vals-1.1 | Professional › Legal | 11.3 | 23.8 | +12.5 |
| DAYJOB: Law (November 2026 release) | Professional › Legal | 9.7 | 20.6 | +10.9 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 46.3 | 25.1 | −10.6 |
| Artificial Analysis Engineering Index | Science › Engineering | 60.1 | 49.6 | −10.5 |
| Artificial Analysis Legal Index | Professional › Legal | 56.7 | 46.9 | −9.9 |
| GBENCH One-shot Coding GScore | Games | 66.8 | 58.5 | −8.3 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 17.4 | 1.5 | −8.0 |
| Artificial Analysis Coding Agent Index v2.1 | Software › Software engineering | 52.3 | 44.5 | −7.8 |
| LMArena Code Arena WebDev | Software › Frontend & design | 1703 | 1648 | −7.3 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 78.4 | 71.2 | −7.3 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 24.0 | 31.2 | +7.2 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 78.0 | 70.9 | −7.1 |
| Artificial Analysis Finance & Accounting Index | Professional › Finance, accounting & tax | 58.1 | 51.1 | −6.9 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 15.7 | 2.5 | −6.6 |
| Professional Reasoning Benchmark - Finance | Professional › Finance, accounting & tax | 52.3 | 58.7 | +6.4 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 74.4 | 68.1 | −6.3 |
| LiveBench 2.0 (2026-11-06 release) | Composite indexes | 77.0 | 70.7 | −6.3 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1497 | 1610 | +6.2 |
| FrontierMath Tiers 1-3 (v2) - Private | Math | 88.1 | 82.1 | −6.0 |
| micro1 PharmaBench | Professional › General knowledge work | 11.4 | 17.2 | +5.8 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 73.7 | 67.9 | −5.8 |
| Artificial Analysis Economics Index | Composite indexes | 55.0 | 49.5 | −5.5 |
| Professional Reasoning Benchmark - Legal | Professional › Legal | 53.1 | 58.4 | +5.3 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 27.2 | 22.2 | −5.0 |
| Agents' Last Exam v1.1 (ALE v1.1) | Professional › General knowledge work | 43.4 | 38.5 | −4.9 |
| Vals Finance Index | Professional › Finance, accounting & tax | 66.1 | 61.2 | −4.9 |
| Vibe Code Bench v1.1 | Software › Frontend & design | 87.8 | 83.1 | −4.7 |
| Artificial Analysis Healthcare & Medical Index | Health & medicine | 53.3 | 48.7 | −4.6 |
| Terminal-Bench 4.0 | Software › Software engineering | 54.0 | 49.5 | −4.5 |
| Epoch Capabilities Index (ECI) | Composite indexes | 167 | 162 | −4.1 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | -4.5 | 3.4 | +4.0 |
| CursorBench 4.0 | Software › Software engineering | 54.3 | 50.4 | −3.9 |
| GradeBench | Professional › Education | 51.9 | 48.1 | −3.8 |
| FORTRESS Average Risk Score (ARS) | Safety & alignment | 17.5 | 13.8 | +3.7 |
| LMArena Text Arena Elo | Language & writing › General assistance | 1490 | 1511 | +3.6 |
| Vals Index v2.2 | Composite indexes | 68.5 | 64.9 | −3.6 |
| Legal Research Bench | Professional › Legal | 53.8 | 50.5 | −3.4 |
| Furniture Assembly Benchmark | Perception & media › Spatial & visual reasoning | 57.5 | 54.2 | −3.3 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 75.8 | 72.5 | −3.3 |
| Artificial Analysis Strategy & Ops Index | Professional › General knowledge work | 57.7 | 54.5 | −3.3 |
| SWE-Marathon v1.1 | Software › Software engineering | 49.4 | 52.5 | +3.1 |
| DAYJOB: Accounting | Professional › Finance, accounting & tax | 33.6 | 30.8 | −2.8 |
| BuseyBench SVG | Perception & media › Generation | 6.9 | 6.6 | −2.6 |
| DAYJOB: Tax | Professional › Finance, accounting & tax | 3.6 | 6.0 | +2.4 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 41.2 | 43.6 | +2.3 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 32.8 | 30.6 | −2.2 |
| CWE-bench v2 Pass@1 | Software › Cybersecurity | 68.1 | 66.1 | −2.0 |
| Vals Legal Index | Professional › Legal | 73.6 | 75.3 | +1.6 |
| Bug Hunt Bench | Software › Software engineering | 31.1 | 30.2 | −1.0 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 54.2 | 54.8 | +0.6 |
| OTIS Mock AIME 2024-2025 | Math | 100.0 | 99.4 | −0.6 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 0.7 | -0.2 | −0.4 |
| Finance Agent v2.1 | Professional › Finance, accounting & tax | 65.1 | 65.5 | +0.4 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 38.5 | 38.8 | +0.3 |
| Dual-Arm Sim (task completion) | Tool & computer use › Robotics | 0.0 | 0.0 | 0.0 |
| DualEntry Accounting AI Benchmark | Professional › Finance, accounting & tax | 83.2 | 83.2 | 0.0 |
Data: Artificial Analysis, Vals AI, Surge AI and 18 othersatlas.village.dev