Gemini 4 Hydrogen vs GPT-6.1 Sol
GPT-6.1 Sol leads by 0.5 capability points (≈95% interval −2.1 to +3.1): statistically tied · P(higher) ≈ 64%
Gemini 4 Hydrogen wins 38 of 73 shared benchmarks (31 lost, 4 tied)
Capability vs. relative cost
Gemini 4 HydrogenGPT-6.1 SolClosed weightsOpen weights
Benchmark wins38 – 31 · 73 shared, 4 tied
Comparing matching setups only (11 benchmarks excluded) · suites counted once
Area← Gemini 4 HydrogenGPT-6.1 Sol →
Data: Artificial Analysis, Vals AI, Epoch AI and 19 othersatlas.village.dev
Margins
Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is GPT-6.1 Sol minus Gemini 4 Hydrogen.
| TxBench-Antibody-Development V1 (60 evaluations) | Science › Life sciences | 60.6 | 27.8 | −32.8 |
| Furniture Assembly Benchmark | Perception & media › Spatial & visual reasoning | 57.5 | 80.0 | +22.5 |
| DAYJOB: Law (November 2026 release) | Professional › Legal | 9.7 | 31.6 | +21.9 |
| GDP.xlsx | Professional › General knowledge work | 40.9 | 22.0 | −18.9 |
| Ramp SWE-Bench v2 (Bash harness) | Software › Software engineering | 70.8 | 52.0 | −18.8 |
| Humanity's Sixth Sense | Perception & media › Spatial & visual reasoning | 63.4 | 46.6 | −16.8 |
| EBR-bench v4 (card ban, single agent) | Games | 70.5 | 54.3 | −16.2 |
| SpireClimb | Games | 7.7 | 23.8 | +16.2 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 78.4 | 62.3 | −16.1 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 44.6 | 28.6 | −16.0 |
| Legal Research Bench | Professional › Legal | 53.8 | 38.5 | −15.4 |
| Mystery Game Puzzles | Games | 65.0 | 80.0 | +15.0 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 41.2 | 54.6 | +13.4 |
| FrontierMath Tier 4 (v2) - Private | Math | 87.8 | 100.0 | +12.2 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 27.2 | 38.3 | +11.1 |
| BuseyBench SVG | Perception & media › Generation | 6.9 | 8.0 | +10.9 |
| GBENCH One-shot Coding GScore | Games | 66.8 | 77.5 | +10.7 |
| Vals Finance Index | Professional › Finance, accounting & tax | 66.1 | 55.8 | −10.3 |
| SimpleBench | Knowledge & reasoning › General knowledge | 91.7 | 82.9 | −8.8 |
| Gray Swan IPI Attack Success (k=15) | Safety & alignment | 0.0 | 8.6 | −8.6 |
| DAYJOB: Accounting | Professional › Finance, accounting & tax | 33.6 | 25.2 | −8.4 |
| Chess Puzzles (task version 2.0) | Games | 60.0 | 68.0 | +8.0 |
| TaxCalcBench TY25 | Professional › Finance, accounting & tax | 44.0 | 36.0 | −8.0 |
| TxBench Antibody Discovery V1 (100 evaluations) | Science › Life sciences | 57.7 | 49.7 | −8.0 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 74.4 | 66.6 | −7.9 |
| Code Migration v1.1 (Vals Index v2.2 subset) | Software › Software engineering | 82.7 | 75.1 | −7.5 |
| HiL-Bench (Human-in-Loop Benchmark) | Tool & computer use › General agents | 55.3 | 62.3 | +7.0 |
| Professional Reasoning Benchmark - Finance | Professional › Finance, accounting & tax | 52.3 | 46.2 | −6.1 |
| FrontierMath Tiers 1-3 (v2) - Private | Math | 88.1 | 93.7 | +5.6 |
| Professional Reasoning Benchmark - Legal | Professional › Legal | 53.1 | 47.9 | −5.2 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 38.5 | 33.5 | −5.0 |
| HLAB-Vals-1.1 | Professional › Legal | 11.3 | 6.3 | −5.0 |
| Vals Legal Index | Professional › Legal | 73.6 | 68.7 | −4.9 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | -4.5 | 5.1 | +4.8 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 75.8 | 71.1 | −4.7 |
| Vals Index v2.2 | Composite indexes | 68.5 | 63.9 | −4.5 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 13.5 | 11.3 | −4.3 |
| LiveBench 2.0 (2026-11-06 release) | Composite indexes | 77.0 | 81.2 | +4.2 |
| Terminal-Bench 4.0 | Software › Software engineering | 54.0 | 58.2 | +4.1 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 73.7 | 70.1 | −3.6 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1569 | 1504 | −3.5 |
| Finance Agent v2.1 | Professional › Finance, accounting & tax | 65.1 | 61.8 | −3.3 |
| SimpleQA Verified (Epoch anti-abstention) | Knowledge & reasoning › General knowledge | 70.9 | 73.9 | +3.0 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 17.4 | 11.8 | −2.8 |
| FORTRESS Average Risk Score (ARS) | Safety & alignment | 17.5 | 14.7 | +2.8 |
| BigFinanceBench private holdout (final-answer accuracy) | Professional › Finance, accounting & tax | 49.4 | 52.0 | +2.6 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 54.2 | 51.6 | −2.6 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 29.3 | 24.5 | −2.4 |
| sudo L7 (October 2026 rescoring) | Software › Software engineering | 31.7 | 33.9 | +2.2 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 15.7 | 11.6 | −2.1 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 80.6 | 82.6 | +2.0 |
| DualEntry Accounting AI Benchmark | Professional › Finance, accounting & tax | 83.2 | 81.2 | −2.0 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 46.3 | 42.4 | −1.9 |
| TxBench Oligo Discovery V1 (113 evaluations) | Science › Life sciences | 54.3 | 52.5 | −1.8 |
| ARC-AGI-2 | Knowledge & reasoning › Abstract reasoning | 92.5 | 94.2 | +1.7 |
| TxBench Preclinical Pharmacology V1 Clean Reruns | Science › Life sciences | 61.3 | 59.7 | −1.7 |
| Epoch Capabilities Index (ECI) | Composite indexes | 167 | 169 | +1.4 |
| MultiChallenge | Language & writing › Conversation | 80.6 | 82.0 | +1.4 |
| GradeBench | Professional › Education | 51.9 | 53.2 | +1.3 |
| LMArena Text Arena Elo | Language & writing › General assistance | 1490 | 1483 | −1.2 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 24.0 | 25.2 | +1.2 |
| Vibe Code Bench v1.1 | Software › Frontend & design | 87.8 | 88.9 | +1.2 |
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 49.5 | 50.5 | +1.0 |
| Agents' Last Exam v1.1 Score | Professional › General knowledge work | 71.6 | 72.5 | +0.9 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 32.8 | 32.0 | −0.8 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 53.5 | 54.0 | +0.5 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 39.0 | 38.5 | −0.4 |
| DAYJOB: Tax | Professional › Finance, accounting & tax | 3.6 | 4.0 | +0.4 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 0.7 | 1.2 | +0.2 |
| ARC-AGI-1 Semi-Private | Knowledge & reasoning › Abstract reasoning | 98.5 | 98.5 | 0.0 |
| Dual-Arm Sim (task completion) | Tool & computer use › Robotics | 0.0 | 0.0 | 0.0 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 58.8 | 58.8 | 0.0 |
| OTIS Mock AIME 2024-2025 | Math | 100.0 | 100.0 | 0.0 |
Data: Artificial Analysis, Vals AI, Epoch AI and 19 othersatlas.village.dev