Atlas

Compare Models

Gemini 4 Neon vs GPT-6.1 Terra

Gemini 4 Neon leads by 3.7 capability points (≈95% interval +0.9 to +6.6): a clear lead over GPT-6.1 Terra · P(higher) ≈ 100%

Wins 47 of 50 shared benchmarks (2 lost, 1 tied)

Capability vs. relative cost
Gemini 4 NeonGPT-6.1 TerraClosed weightsOpen weights

Benchmark wins47 – 2 · 50 shared, 1 tied

Comparing matching setups only (31 benchmarks excluded) · suites counted once
← Gemini 4 NeonGPT-6.1 Terra →
Data: Artificial Analysis, Vals AI, Epoch AI and 6 othersatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is GPT-6.1 Terra minus Gemini 4 Neon.

Vending-Bench 2Business simulations168439107−45.8
Legal Research BenchProfessional › Legal56.326.0−30.3
Terminal-Bench-Science 0.1Science › Multidisciplinary70.041.9−28.1
Tax Agent BenchProfessional › Finance, accounting & tax79.855.0−24.8
Humanity's Last Exam DiamondKnowledge & reasoning › General knowledge71.347.3−24.0
τ³-Banking (Mercor run, bm25_grep retrieval)Tool & computer use › Customer service & operations59.537.1−22.4
SimpleBenchKnowledge & reasoning › General knowledge94.472.2−22.2
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Professional › Legal26.64.5−22.1
AA-Briefcase v1.1 EloProfessional › General knowledge work16991334−20.0
Vals Legal IndexProfessional › Legal78.458.8−19.6
AutomationBench-AA 1.0.6Tool & computer use › Customer service & operations79.360.2−19.2
HLAB-Vals-1.1Professional › Legal19.61.7−17.9
SimpleQA Verified (Epoch anti-abstention)Knowledge & reasoning › General knowledge75.257.3−17.9
Artificial Analysis Legal IndexProfessional › Legal63.846.4−17.4
Terminal-Bench 4.0 (Artificial Analysis)Software › Software engineering63.646.5−17.2
Artificial Analysis Finance & Accounting IndexProfessional › Finance, accounting & tax65.149.0−16.1
LMArena Agent Arena Net Improvement (Steering Burden)Tool & computer use › General agents18.410.5−15.5
Artificial Analysis Healthcare & Medical IndexHealth & medicine60.144.7−15.4
Vals Finance IndexProfessional › Finance, accounting & tax68.353.3−15.0
Artificial Analysis Strategy & Ops IndexProfessional › General knowledge work63.850.0−13.8
Artificial Analysis Economics IndexComposite indexes61.948.5−13.4
LMArena Agent Arena Praise vs ComplaintTool & computer use › General agents44.517.9−13.3
MKA (Artificial Analysis)Professional › General knowledge work56.543.8−12.7
Finance Agent v2.1Professional › Finance, accounting & tax68.656.0−12.6
AA-Omniscience v1.1 IndexKnowledge & reasoning › General knowledge54.829.6−12.6
Terminal-Bench 4.0Software › Software engineering61.649.5−12.1
Artificial Analysis Engineering IndexScience › Engineering66.854.7−12.1
Mystery Game PuzzlesGames85.074.0−11.0
Vals Index v2.2Composite indexes70.960.5−10.3
ChartographyPerception & media › Documents & charts73.563.2−10.3
Artificial Analysis Intelligence Index v5.0.1Composite indexes61.451.3−10.1
Furniture Assembly BenchmarkPerception & media › Spatial & visual reasoning89.279.2−10.0
Vibe Code Bench v1.1Software › Frontend & design93.784.0−9.7
APEX-Agents 1.1Professional › General knowledge work88.979.3−9.6
LMArena Agent Arena Bash ReliabilityTool & computer use › General agents-6.012.7+9.3
LiveBench 2.0 (2026-11-06 release)Composite indexes87.478.6−8.9
FrontierCode 2.0 (Artificial Analysis)Software › Software engineering44.436.9−7.6
APEX-Accounting 1.1Professional › Finance, accounting & tax70.863.7−7.1
LMArena Agent Arena Confirmed SuccessTool & computer use › General agents19.28.5−5.3
FrontierMath Tier 4 (v2) - PrivateMath97.692.7−4.9
GDP.pdf (Artificial Analysis, All-pass)Professional › General knowledge work33.428.6−4.8
LMArena Agent Arena Steering BurdenTool & computer use › General agents21.213.2−4.0
Code Migration v1.1 (Vals Index v2.2 subset)Software › Software engineering75.872.2−3.6
Excel Modeling Benchmark v1.1Professional › Finance, accounting & tax76.573.3−3.1
GPQA DiamondScience › Multidisciplinary96.394.3−2.0
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & medicine31.730.6−1.1
FrontierMath Tiers 1-3 (v2) - PrivateMath91.290.9−0.4
CritPt v2 (agentic, avg@5)Science › Physical sciences77.477.1−0.3
LMArena Agent Arena Tool HallucinationTool & computer use › Tool calling1.01.2+0.1
OTIS Mock AIME 2024-2025Math100.0100.00.0
Data: Artificial Analysis, Vals AI, Epoch AI and 6 othersatlas.village.dev
Win counts use each benchmark’s declared score direction. Represented suite slices and the derived Capability score are excluded from headline totals. Missing or incompatible setups are excluded, never counted as losses.
Loading Atlas data…