Atlas

Compare Models

Grok 4.7 vs Muse Ember

Muse Ember leads by 1.6 capability points (≈95% interval −0.7 to +3.9): statistically tied · P(higher) ≈ 91%

Grok 4.7 wins 36 of 61 shared benchmarks (24 lost, 1 tied)

Capability vs. relative cost
Grok 4.7Muse EmberClosed weightsOpen weights

Benchmark wins36 – 24 · 61 shared, 1 tied

Comparing matching setups only (3 benchmarks excluded) · suites counted once
← Grok 4.7Muse Ember →
Data: Artificial Analysis, Vals AI, Surge AI and 10 othersatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Muse Ember minus Grok 4.7.

DAXBench v2Software › Data & ML92.662.8−29.8
τ³-Banking (Mercor run, bm25_grep retrieval)Tool & computer use › Customer service & operations32.653.3+20.7
HANDBOOK.md Agents v1.1Tool & computer use › Customer service & operations44.029.8−14.2
Ramp SWE-Bench v2Software › Software engineering64.451.8−12.6
AA-Omniscience v1.1 IndexKnowledge & reasoning › General knowledge31.27.7−11.7
Terminal-Bench 4.0Software › Software engineering28.839.4+10.6
CritPt v2 (agentic, avg@5)Science › Physical sciences51.462.0+10.6
DAYJOB: Healthcare (October 2026 rescoring)Health & medicine27.617.2−10.4
Terminal-Bench 4.0 (Artificial Analysis)Software › Software engineering25.835.4+9.6
Artificial Analysis Legal IndexProfessional › Legal47.238.5−8.7
BuseyBench SVGPerception & media › Generation4.65.4+8.3
Artificial Analysis Finance & Accounting IndexProfessional › Finance, accounting & tax51.744.6−7.1
GDP.xlsxProfessional › General knowledge work22.315.4−6.9
FrontierCode 2.0 (Artificial Analysis)Software › Software engineering28.722.0−6.7
HLAB-Vals-1.1Professional › Legal18.325.0+6.7
Terminal-Bench-Science 0.1Science › Multidisciplinary11.417.6+6.2
Artificial Analysis Strategy & Ops IndexProfessional › General knowledge work54.848.7−6.1
SWE Atlas v1.1 - OverallSoftware › Software engineering62.156.0−6.1
EnterpriseBench: CoreCraftTool & computer use › Customer service & operations68.262.1−6.1
TaxCalcBench TY25Professional › Finance, accounting & tax8.02.0−6.0
AA-Briefcase v1.1 EloProfessional › General knowledge work16341526−5.9
FORTRESS Average Risk Score (ARS)Safety & alignment6.311.9−5.6
Artificial Analysis Healthcare & Medical IndexHealth & medicine46.641.3−5.3
APEX-1 (Mercor, mean score)Professional › General knowledge work65.470.4+5.0
Artificial Analysis Economics IndexComposite indexes45.741.0−4.7
Code Migration v1.1 (Vals Index v2.2 subset)Software › Software engineering51.847.1−4.7
APEX-Agents 1.1Professional › General knowledge work80.475.7−4.7
Chartography (November 2026 rescoring)Perception & media › Documents & charts74.679.3+4.7
LMArena Agent Arena Steering BurdenTool & computer use › General agents-4.34.8+4.5
GBENCH One-shot Coding GScoreGames60.556.4−4.1
LMArena Code Arena WebDevSoftware › Frontend & design16391608−4.1
RiemannBench (November 2026 rescoring)Math57.653.6−4.0
Finance Agent v2.1Professional › Finance, accounting & tax60.663.8+3.2
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Professional › Legal19.522.7+3.2
Vals Legal IndexProfessional › Legal72.775.7+3.0
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & medicine21.118.3−2.8
LMArena Agent Arena Confirmed SuccessTool & computer use › General agents3.5-2.0−2.8
AutomationBench-AA 1.0.6Tool & computer use › Customer service & operations65.668.1+2.5
Vibe Code Bench v1.1Software › Frontend & design86.283.7−2.5
LMArena Agent Arena Praise vs ComplaintTool & computer use › General agents2.1-2.7−2.4
DeepFinanceProfessional › Finance, accounting & tax34.136.5+2.4
Tax Agent BenchProfessional › Finance, accounting & tax69.071.4+2.4
DAYJOB: Finance (October 2026 rescoring)Professional › Finance, accounting & tax26.524.2−2.3
Artificial Analysis Progress Index v1Composite indexes323329+2.1
Vals Finance IndexProfessional › Finance, accounting & tax58.860.9+2.1
ComplexConstraints (November 2026 rescoring)Language & writing › Instruction following60.858.8−2.0
Artificial Analysis Coding Agent Index v2.1Software › Software engineering41.639.7−2.0
Legal Research BenchProfessional › Legal47.149.0+1.9
Vals Index v2.2Composite indexes58.660.5+1.9
Humanity's Last Exam DiamondKnowledge & reasoning › General knowledge23.625.1+1.5
LMArena Agent Arena Net Improvement (Steering Burden)Tool & computer use › General agents1.51.0−1.0
LMArena Agent Arena Bash ReliabilityTool & computer use › General agents8.26.6−0.8
Excel Modeling Benchmark v1.1Professional › Finance, accounting & tax71.572.2+0.7
Artificial Analysis Coding Agent Index v2.0Software › Software engineering46.645.9−0.7
Artificial Analysis Intelligence Index v5.0.1Composite indexes35.536.0+0.5
APEX-SWESoftware › Software engineering53.452.9−0.5
GDP.pdf (November 2026 rescoring)Professional › General knowledge work48.849.2+0.4
GDP.pdf (Artificial Analysis, All-pass)Professional › General knowledge work23.222.8−0.4
LMArena Agent Arena Tool HallucinationTool & computer use › Tool calling1.20.9−0.1
Artificial Analysis Engineering IndexScience › Engineering41.841.80.0
FrontierCUA v1 pass@1Tool & computer use › Computer & browser use7.57.50.0
Data: Artificial Analysis, Vals AI, Surge AI 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…