Gemini 4 Argon vs Gemini 4 Neon
TryTop 2 now
Gemini 4 Neon leads by 3.3 capability points (≈95% interval +0.5 to +6.1): a clear lead over Gemini 4 Argon · P(higher) ≈ 99%
Wins 61 of 74 shared benchmarks (12 lost, 1 tied)
Capability vs. relative cost
Gemini 4 ArgonGemini 4 NeonClosed weightsOpen weights
Benchmark wins12 – 61 · 74 shared, 1 tied
Comparing matching setups only · suites counted once
Area← Gemini 4 ArgonGemini 4 Neon →
Data: Vals AI, Artificial Analysis, Surge AI and 14 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 Argon.
| DAYJOB: Accounting | Professional › Finance, accounting & tax | 18.0 | 49.6 | +31.6 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 42.5 | 71.3 | +28.8 |
| Chess Puzzles (task version 2.0) | Games | 59.0 | 83.0 | +24.0 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 23.6 | 47.2 | +23.6 |
| sudo L7 (October 2026 rescoring) | Software › Software engineering | 16.1 | 38.3 | +22.2 |
| Mystery Game Puzzles | Games | 64.0 | 85.0 | +21.0 |
| Humanity's Sixth Sense | Perception & media › Spatial & visual reasoning | 55.4 | 75.9 | +20.5 |
| DAYJOB: Law (November 2026 release) | Professional › Legal | 5.9 | 26.2 | +20.3 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 30.5 | 50.5 | +20.0 |
| FrontierMath Tier 4 (v2) - Private | Math | 78.0 | 97.6 | +19.5 |
| Terminal-Bench Science 0.1 (Vals) | Science › Multidisciplinary | 44.3 | 62.9 | +18.6 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 13.9 | 31.7 | +17.8 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 25.7 | 43.3 | +17.6 |
| Furniture Assembly Benchmark | Perception & media › Spatial & visual reasoning | 72.5 | 89.2 | +16.7 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 10.3 | 18.4 | +15.8 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 36.6 | 52.3 | +15.7 |
| Bug Hunt Bench | Software › Software engineering | 38.7 | 54.3 | +15.6 |
| Artificial Analysis Coding Agent Index v2.1 | Software › Software engineering | 45.5 | 58.6 | +13.1 |
| LMArena Code Arena WebDev | Software › Frontend & design | 1681 | 1764 | +11.1 |
| GraphWalks BFS: 256K to 1M Aggregate | Knowledge & reasoning › Reading & long context | 84.2 | 95.2 | +11.0 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 50.4 | 61.4 | +11.0 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 23.8 | 44.5 | +10.4 |
| LiveBench 2.0 (2026-11-06 release) | Composite indexes | 77.7 | 87.4 | +9.8 |
| DAYJOB: Tax | Professional › Finance, accounting & tax | 3.2 | 12.8 | +9.6 |
| RiemannBench (November 2026 rescoring) | Math | 74.4 | 84.0 | +9.6 |
| CursorBench 4.1 | Software › Software engineering | 56.5 | 65.4 | +8.9 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1485 | 1632 | +8.1 |
| SWE-Marathon v1.2 | Software › Software engineering | 54.5 | 62.5 | +8.0 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 25.6 | 33.4 | +7.8 |
| HLAB-Vals-1.1 | Professional › Legal | 27.1 | 19.6 | −7.5 |
| Vals Science Index | Science › Multidisciplinary | 55.4 | 62.6 | +7.3 |
| SEC-Bench Pro - release-reported cohort | Software › Cybersecurity | 83.4 | 90.5 | +7.1 |
| Agents' Last Exam v1.1 Score | Professional › General knowledge work | 69.0 | 75.8 | +6.8 |
| ExploitGym - release-reported percentage, cohort unspecified | Software › Cybersecurity | 42.0 | 48.3 | +6.3 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 71.7 | 78.0 | +6.3 |
| AutomationBench 1.0.6 (Zapier) | Tool & computer use › Customer service & operations | 50.1 | 56.3 | +6.2 |
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 48.6 | 54.8 | +6.2 |
| CWE-bench v2 Pass@1 | Software › Cybersecurity | 65.0 | 70.7 | +5.7 |
| FrontierMath Tiers 1-3 (v2) - Private | Math | 86.0 | 91.2 | +5.3 |
| GDP.xlsx | Professional › General knowledge work | 38.3 | 43.1 | +4.8 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 12.2 | 21.2 | +4.5 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 46.1 | 54.8 | +4.3 |
| Terminal-Bench 4.0 | Software › Software engineering | 57.6 | 61.6 | +4.0 |
| Vals RSI Index v1.1 (four tasks) | Software › Data & ML | 30.6 | 34.4 | +3.8 |
| BioMysteryBench (Vals Terminus 2) | Science › Life sciences | 76.3 | 80.0 | +3.7 |
| MKA (Artificial Analysis) | Professional › General knowledge work | 51.3 | 54.6 | +3.3 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 12.6 | 19.2 | +3.3 |
| LVBench | Perception & media › Video understanding | 91.7 | 89.0 | −2.7 |
| SimpleBench | Knowledge & reasoning › General knowledge | 91.7 | 94.4 | +2.7 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 70.3 | 72.9 | +2.6 |
| Vals Finance Index | Professional › Finance, accounting & tax | 66.0 | 68.3 | +2.3 |
| APEX-1 (Mercor, mean score) | Professional › General knowledge work | 81.1 | 78.9 | −2.2 |
| Chartography | Perception & media › Documents & charts | 71.6 | 73.5 | +1.9 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 77.5 | 79.3 | +1.8 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 74.8 | 76.5 | +1.6 |
| Vals Legal Index | Professional › Legal | 76.8 | 78.4 | +1.6 |
| Legal Research Bench | Professional › Legal | 54.8 | 56.3 | +1.4 |
| τ³-Banking (Mercor run, bm25_grep retrieval) | Tool & computer use › Customer service & operations | 58.1 | 59.5 | +1.4 |
| APEX-Agents 1.1 | Professional › General knowledge work | 87.4 | 88.8 | +1.4 |
| Code Migration v1.1 (Vals Index v2.2 subset) | Software › Software engineering | 74.4 | 75.8 | +1.3 |
| Finance Agent v2.1 | Professional › Finance, accounting & tax | 67.7 | 68.6 | +0.9 |
| SimpleQA Verified (Epoch anti-abstention) | Knowledge & reasoning › General knowledge | 76.0 | 75.2 | −0.8 |
| Harvey LAB (Mercor run, undisclosed ~213-task cohort) | Professional › Legal | 27.4 | 26.6 | −0.8 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 79.0 | 79.8 | +0.8 |
| APEX-Accounting 1.1 | Professional › Finance, accounting & tax | 71.5 | 70.8 | −0.7 |
| Gray Swan IPI Attack Success (k=15) | Safety & alignment | 0.7 | 0.0 | +0.7 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 55.1 | 54.5 | −0.5 |
| MysteryMechanism | Math | 45.5 | 45.0 | −0.4 |
| LMArena Text Arena Elo | Language & writing › General assistance | 1521 | 1518 | −0.4 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 53.6 | 53.4 | −0.2 |
| GPQA Diamond | Science › Multidisciplinary | 96.4 | 96.3 | −0.1 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 0.8 | 1.0 | +0.1 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | -6.0 | -6.0 | 0.0 |
| OTIS Mock AIME 2024-2025 | Math | 100.0 | 100.0 | 0.0 |
Data: Vals AI, Artificial Analysis, Surge AI and 14 othersatlas.village.dev