Gemini 4 Hydrogen vs Gemini 4 Neon
TryTop 2 now
Gemini 4 Neon leads by 3.3 capability points (≈95% interval +0.4 to +6.1): a clear lead over Gemini 4 Hydrogen · P(higher) ≈ 99%
Wins 59 of 66 shared benchmarks (6 lost, 1 tied)
Capability vs. relative cost
Gemini 4 HydrogenGemini 4 NeonClosed weightsOpen weights
Benchmark wins6 – 59 · 66 shared, 1 tied
Comparing matching setups only (1 benchmark excluded) · suites counted once
Area← Gemini 4 HydrogenGemini 4 Neon →
Data: Artificial Analysis, Vals AI, Surge AI and 11 othersatlas.village.dev
Margins
Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Gemini 4 Neon minus Gemini 4 Hydrogen.
| Furniture Assembly Benchmark | Perception & media › Spatial & visual reasoning | 57.5 | 89.2 | +31.7 |
| Bug Hunt Bench | Software › Software engineering | 31.1 | 60.0 | +28.9 |
| SWE-sweep | Software › Software engineering | 13.5 | 39.0 | +25.5 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 24.0 | 47.2 | +23.2 |
| Chess Puzzles (task version 2.0) | Games | 60.0 | 83.0 | +23.0 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 41.2 | 61.4 | +20.2 |
| Mystery Game Puzzles | Games | 65.0 | 85.0 | +20.0 |
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 51.4 | 70.5 | +19.1 |
| DAYJOB: Law (November 2026 release) | Professional › Legal | 9.7 | 26.2 | +16.5 |
| DAYJOB: Accounting | Professional › Finance, accounting & tax | 33.6 | 49.6 | +16.0 |
| Humanity's Sixth Sense | Perception & media › Spatial & visual reasoning | 63.4 | 75.9 | +12.5 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 58.8 | 71.3 | +12.5 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 38.5 | 50.5 | +12.0 |
| LiveBench 2.0 (2026-11-06 release) | Composite indexes | 77.0 | 87.4 | +10.4 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 53.5 | 63.6 | +10.1 |
| FrontierMath Tier 4 (v2) - Private | Math | 87.8 | 97.6 | +9.8 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 13.5 | 18.4 | +9.5 |
| DAYJOB: Tax | Professional › Finance, accounting & tax | 3.6 | 12.8 | +9.2 |
| HLAB-Vals-1.1 | Professional › Legal | 11.3 | 19.6 | +8.3 |
| LMArena Code Arena WebDev | Software › Frontend & design | 1703 | 1764 | +8.1 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 44.6 | 52.3 | +7.7 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 29.3 | 44.5 | +7.6 |
| Terminal-Bench 4.0 | Software › Software engineering | 54.0 | 61.6 | +7.6 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1569 | 1699 | +7.1 |
| Artificial Analysis Legal Index | Professional › Legal | 56.7 | 63.8 | +7.0 |
| Artificial Analysis Finance & Accounting Index | Professional › Finance, accounting & tax | 58.1 | 65.1 | +7.0 |
| Code Migration v1.1 (Vals Index v2.2 subset) | Software › Software engineering | 82.7 | 75.8 | −6.9 |
| Chartography | Perception & media › Documents & charts | 66.6 | 73.5 | +6.9 |
| Artificial Analysis Economics Index | Composite indexes | 55.0 | 61.9 | +6.9 |
| Artificial Analysis Healthcare & Medical Index | Health & medicine | 53.3 | 60.1 | +6.8 |
| Artificial Analysis Engineering Index | Science › Engineering | 60.1 | 66.8 | +6.7 |
| sudo L7 (October 2026 rescoring) | Software › Software engineering | 31.7 | 38.3 | +6.6 |
| Artificial Analysis Coding Agent Index v2.1 | Software › Software engineering | 52.3 | 58.6 | +6.3 |
| Artificial Analysis Strategy & Ops Index | Professional › General knowledge work | 57.7 | 63.8 | +6.0 |
| Vibe Code Bench v1.1 | Software › Frontend & design | 87.8 | 93.7 | +5.9 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 39.0 | 44.4 | +5.5 |
| AutomationBench 1.0.6 (Zapier) | Tool & computer use › Customer service & operations | 51.1 | 56.3 | +5.2 |
| LMArena Text Arena Elo | Language & writing › General assistance | 1490 | 1518 | +5.0 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 74.4 | 79.3 | +4.9 |
| Vals Legal Index | Professional › Legal | 73.6 | 78.4 | +4.7 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 27.2 | 31.7 | +4.4 |
| SWE-Marathon v1.2 | Software › Software engineering | 58.1 | 62.5 | +4.4 |
| SimpleQA Verified (Epoch anti-abstention) | Knowledge & reasoning › General knowledge | 70.9 | 75.2 | +4.3 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 46.3 | 54.8 | +4.3 |
| Agents' Last Exam v1.1 Score | Professional › General knowledge work | 71.6 | 75.8 | +4.2 |
| CursorBench 4.1 | Software › Software engineering | 61.4 | 65.4 | +4.0 |
| Finance Agent v2.1 | Professional › Finance, accounting & tax | 65.1 | 68.6 | +3.4 |
| FrontierMath Tiers 1-3 (v2) - Private | Math | 88.1 | 91.2 | +3.2 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 80.6 | 77.4 | −3.1 |
| SimpleBench | Knowledge & reasoning › General knowledge | 91.7 | 94.4 | +2.7 |
| CWE-bench v2 Pass@1 | Software › Cybersecurity | 68.1 | 70.7 | +2.6 |
| Legal Research Bench | Professional › Legal | 53.8 | 56.3 | +2.4 |
| Vals Index v2.2 | Composite indexes | 68.5 | 70.9 | +2.4 |
| GDP.xlsx | Professional › General knowledge work | 40.9 | 43.1 | +2.2 |
| Vals Finance Index | Professional › Finance, accounting & tax | 66.1 | 68.3 | +2.2 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 17.4 | 21.2 | +1.9 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 15.7 | 19.2 | +1.7 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 78.4 | 79.8 | +1.4 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 73.7 | 72.9 | −0.8 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 54.2 | 53.4 | −0.8 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | -4.5 | -6.0 | −0.7 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 75.8 | 76.5 | +0.7 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 32.8 | 33.4 | +0.6 |
| GPQA Diamond | Science › Multidisciplinary | 96.8 | 96.3 | −0.5 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 0.7 | 1.0 | +0.1 |
| OTIS Mock AIME 2024-2025 | Math | 100.0 | 100.0 | 0.0 |
Data: Artificial Analysis, Vals AI, Surge AI and 11 othersatlas.village.dev