Atlas

Compare Models

Gemini 4 Neon vs Muse Spark 1.3

Gemini 4 Neon leads by 7.9 capability points (≈95% interval +5.3 to +10.6): a clear lead over Muse Spark 1.3 · P(higher) ≈ 100%

Wins 53 of 55 shared benchmarks (2 lost)

Capability vs. relative cost
Gemini 4 NeonMuse Spark 1.3Closed weightsOpen weights

Benchmark wins53 – 2 · 55 shared

Comparing matching setups only (4 benchmarks excluded) · suites counted once
← Gemini 4 NeonMuse Spark 1.3 →
Data: Surge AI, Artificial Analysis, Arena and 10 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 Neon.

Terminal-Bench-Science 0.1Science › Multidisciplinary70.03.8−66.2
Mystery Game PuzzlesGames85.025.0−60.0
Terminal-Bench Science 0.1 (Vals)Science › Multidisciplinary62.94.3−58.6
FrontierMath Tier 4 (v2) - PrivateMath97.646.3−51.2
Terminal-Bench 4.0Software › Software engineering61.610.6−51.0
Terminal-Bench 4.0 (Artificial Analysis)Software › Software engineering63.616.7−47.0
Chess Puzzles (task version 2.0)Games83.037.0−46.0
ChartographyPerception & media › Documents & charts73.527.6−45.9
Humanity's Sixth SensePerception & media › Spatial & visual reasoning75.937.4−38.5
Humanity's Last Exam DiamondKnowledge & reasoning › General knowledge71.333.7−37.6
AutomationBench 1.0.6 (Zapier)Tool & computer use › Customer service & operations56.320.7−35.6
LMArena Agent Arena Net Improvement (Steering Burden)Tool & computer use › General agents18.41.1−33.7
DAYJOB: AccountingProfessional › Finance, accounting & tax49.618.0−31.6
RiemannBench (November 2026 rescoring)Math84.053.6−30.4
Bug Hunt BenchSoftware › Software engineering60.030.7−29.3
sudo L7 (October 2026 rescoring)Software › Software engineering38.312.8−25.5
DAYJOB: Healthcare (October 2026 rescoring)Health & medicine47.222.0−25.2
Artificial Analysis Intelligence Index v5.0.1Composite indexes61.436.6−24.8
CursorBench 4.1Software › Software engineering65.441.9−23.5
DAYJOB: Finance (October 2026 rescoring)Professional › Finance, accounting & tax50.527.0−23.5
LMArena Agent Arena Praise vs ComplaintTool & computer use › General agents44.50.1−22.2
Artificial Analysis Coding Agent Index v2.1Software › Software engineering58.636.8−21.9
AutomationBench-AA 1.0.6Tool & computer use › Customer service & operations79.357.9−21.5
FrontierCode 2.0 (Artificial Analysis)Software › Software engineering44.423.1−21.3
GDP.xlsxProfessional › General knowledge work43.122.6−20.5
LiveBench 2.0 (2026-11-06 release)Composite indexes87.468.6−18.8
APEX-Agents 1.1Professional › General knowledge work88.970.4−18.5
HANDBOOK.md Agents v1.1Tool & computer use › Customer service & operations52.333.9−18.4
FrontierMath Tiers 1-3 (v2) - PrivateMath91.274.4−16.8
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & medicine31.715.6−16.1
EnterpriseBench: CoreCraftTool & computer use › Customer service & operations77.461.9−15.5
AA-Omniscience v1.1 IndexKnowledge & reasoning › General knowledge54.823.8−15.5
Vals RSI Index v1.1 (four tasks)Software › Data & ML34.419.6−14.8
LMArena Code Arena WebDevSoftware › Frontend & design17641654−14.6
DAYJOB: Law (November 2026 release)Professional › Legal26.211.8−14.4
SimpleBenchKnowledge & reasoning › General knowledge94.481.8−12.6
AA-Briefcase v1.1 EloProfessional › General knowledge work16991480−12.0
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Professional › Legal26.615.0−11.6
Vibe Code Bench v1.1Software › Frontend & design93.782.9−10.8
LMArena Agent Arena Steering BurdenTool & computer use › General agents21.20.8−10.2
CritPt v2 (agentic, avg@5)Science › Physical sciences78.068.0−10.0
DAYJOB: TaxProfessional › Finance, accounting & tax12.83.6−9.2
MysteryMechanismMath45.036.0−9.0
LMArena Agent Arena Confirmed SuccessTool & computer use › General agents19.22.5−8.4
ComplexConstraints (November 2026 rescoring)Language & writing › Instruction following72.964.8−8.1
τ³-Banking (Mercor run, bm25_grep retrieval)Tool & computer use › Customer service & operations59.552.2−7.3
GDP.pdf (Artificial Analysis, All-pass)Professional › General knowledge work33.426.6−6.8
Tax Agent BenchProfessional › Finance, accounting & tax79.873.4−6.4
APEX-Accounting 1.1Professional › Finance, accounting & tax70.864.9−5.9
LMArena Text Arena EloLanguage & writing › General assistance15181498−3.6
LMArena Agent Arena Bash ReliabilityTool & computer use › General agents-6.01.2+3.6
Legal Research BenchProfessional › Legal56.355.3−1.0
OTIS Mock AIME 2024-2025Math100.099.2−0.8
GDP.pdf (November 2026 rescoring)Professional › General knowledge work53.453.0−0.4
LMArena Agent Arena Tool HallucinationTool & computer use › Tool calling1.01.00.0
Data: Surge AI, Artificial Analysis, Arena and 10 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…