Gemini 4 Hydrogen vs Muse Spark 1.3
Gemini 4 Hydrogen leads by 4.7 capability points (≈95% interval +2.3 to +7.0): a clear lead over Muse Spark 1.3 · P(higher) ≈ 100%
Wins 50 of 60 shared benchmarks (9 lost, 1 tied)
Capability vs. relative cost
Gemini 4 HydrogenMuse Spark 1.3Closed weightsOpen weights
Benchmark wins50 – 9 · 60 shared, 1 tied
Comparing matching setups only (2 benchmarks excluded) · suites counted once
Area← Gemini 4 HydrogenMuse Spark 1.3 →
Data: Artificial Analysis, Surge AI, Arena and 17 othersatlas.village.dev
Margins
Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Muse Spark 1.3 minus Gemini 4 Hydrogen.
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 49.5 | 3.8 | −45.7 |
| Terminal-Bench 4.0 | Software › Software engineering | 54.0 | 10.6 | −43.4 |
| FrontierMath Tier 4 (v2) - Private | Math | 87.8 | 46.3 | −41.5 |
| Mystery Game Puzzles | Games | 65.0 | 25.0 | −40.0 |
| Chartography | Perception & media › Documents & charts | 66.6 | 27.6 | −39.0 |
| TaxCalcBench TY25 | Professional › Finance, accounting & tax | 44.0 | 6.0 | −38.0 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 53.5 | 16.7 | −36.9 |
| AutomationBench 1.0.6 (Zapier) | Tool & computer use › Customer service & operations | 51.1 | 20.7 | −30.4 |
| FrontierSWE V2 | Software › Software engineering | 53.1 | 25.8 | −27.3 |
| Humanity's Sixth Sense | Perception & media › Spatial & visual reasoning | 63.4 | 37.4 | −26.0 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 58.8 | 33.7 | −25.1 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 13.5 | 1.1 | −24.2 |
| Chess Puzzles (task version 2.0) | Games | 60.0 | 37.0 | −23.0 |
| CursorBench 4.1 | Software › Software engineering | 61.4 | 41.9 | −19.5 |
| sudo L7 (October 2026 rescoring) | Software › Software engineering | 31.7 | 12.8 | −18.9 |
| GDP.xlsx | Professional › General knowledge work | 40.9 | 22.6 | −18.3 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 74.4 | 57.9 | −16.6 |
| Gray Swan IPI Attack Success (k=15) | Safety & alignment | 0.0 | 15.9 | −15.9 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 39.0 | 23.1 | −15.8 |
| DAYJOB: Accounting | Professional › Finance, accounting & tax | 33.6 | 18.0 | −15.6 |
| Artificial Analysis Coding Agent Index v2.1 | Software › Software engineering | 52.3 | 36.8 | −15.5 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 29.3 | 0.1 | −14.6 |
| FrontierMath Tiers 1-3 (v2) - Private | Math | 88.1 | 74.4 | −13.7 |
| GBENCH One-shot Coding GScore | Games | 66.8 | 53.7 | −13.1 |
| CursorBench 4.0 | Software › Software engineering | 54.3 | 41.6 | −12.7 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 80.6 | 68.0 | −12.6 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 27.2 | 15.6 | −11.7 |
| HiL-Bench (Human-in-Loop Benchmark) | Tool & computer use › General agents | 55.3 | 66.9 | +11.6 |
| BuseyBench SVG | Perception & media › Generation | 6.9 | 5.7 | −11.5 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 38.5 | 27.0 | −11.5 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 46.3 | 23.8 | −11.2 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 44.6 | 33.9 | −10.7 |
| GradeBench | Professional › Education | 51.9 | 41.4 | −10.5 |
| SimpleBench | Knowledge & reasoning › General knowledge | 91.7 | 81.8 | −9.9 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 73.7 | 64.8 | −8.9 |
| Professional Reasoning Benchmark - Legal | Professional › Legal | 53.1 | 61.6 | +8.5 |
| LiveBench 2.0 (2026-11-06 release) | Composite indexes | 77.0 | 68.6 | −8.4 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 17.4 | 0.8 | −8.3 |
| Professional Reasoning Benchmark - Finance | Professional › Finance, accounting & tax | 52.3 | 59.5 | +7.3 |
| DrugDiscoveryBench | Science › Life sciences | 69.1 | 62.2 | −6.9 |
| Epoch Capabilities Index (ECI) | Composite indexes | 167 | 158 | −6.9 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 15.7 | 2.5 | −6.6 |
| LMArena Code Arena WebDev | Software › Frontend & design | 1703 | 1654 | −6.5 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 32.8 | 26.6 | −6.2 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 78.4 | 73.4 | −5.1 |
| Vibe Code Bench v1.1 | Software › Frontend & design | 87.8 | 82.9 | −4.9 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1569 | 1480 | −4.9 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 41.2 | 36.6 | −4.6 |
| SWE-Marathon v1.1 | Software › Software engineering | 49.4 | 46.3 | −3.1 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | -4.5 | 1.2 | +2.8 |
| micro1 PharmaBench | Professional › General knowledge work | 11.4 | 13.9 | +2.5 |
| DAYJOB: Law (November 2026 release) | Professional › Legal | 9.7 | 11.8 | +2.1 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 24.0 | 22.0 | −2.0 |
| Legal Research Bench | Professional › Legal | 53.8 | 55.3 | +1.4 |
| LMArena Text Arena Elo | Language & writing › General assistance | 1490 | 1498 | +1.4 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 54.2 | 53.0 | −1.2 |
| OTIS Mock AIME 2024-2025 | Math | 100.0 | 99.2 | −0.8 |
| Bug Hunt Bench | Software › Software engineering | 31.1 | 30.7 | −0.4 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 0.7 | 1.0 | +0.2 |
| DAYJOB: Tax | Professional › Finance, accounting & tax | 3.6 | 3.6 | 0.0 |
Data: Artificial Analysis, Surge AI, Arena and 17 othersatlas.village.dev