AA-AnalystAgent ↗41.25%
Grok 4.6 (High) · AA-AnalystAgent · strict score ↗
xAI / MODEL EVIDENCE
Grok 4.6
56 benchmarks · 86 observations · 60 sources · released 2026-08-12
Scores at a glance
56 benchmarks · large dot: Grok 4.6; small dots: closest models where reported
EnterpriseOps-Gym-AA ↗48.34%
Grok 4.6 (High) · EnterpriseOps-Gym-AA ↗
GPQA Diamond ↗94.7 %
Grok 4.6 · GPQA Diamond · Accuracy ↗
3 more reported results
94.95% · Grok 4.6 (High) · GPQA Diamond · AA implementation · Artificial Analysis ↗
94% · Grok 4.6 (high) · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗
93.18% · Grok 4.6 (xhigh) · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗
τ³-Banking ↗50.72%
Grok 4.6 (High) · τ³-Banking ↗
Terminal-Bench 2.1 ↗88.39%
Grok 4.6 (High) · Terminal-Bench 2.1 ↗
Grok 4.6 (xhigh) · FrontierMath-Tiers-1-3-v2-Private · Best score (across scorers) · % ↗
Grok 4.6 (xhigh) · FrontierMath-Tier-4-v2-Private · Best score (across scorers) · % ↗
Grok 4.6 (xhigh) · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % ↗
1 more reported result
97.78% · Grok 4.6 (high) · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % · Epoch AI · OTIS Mock AIME 2024-2025 ↗
SimpleQA Verified ↗49.3%
Grok 4.6 (high) · SimpleQA Verified · Best score (across scorers) · % ↗
1 more reported result
48.9% · Grok 4.6 (xhigh) · SimpleQA Verified · Best score (across scorers) · % · Epoch AI · SimpleQA Verified ↗
Grok 4.6 (xhigh) · Chess Puzzles · Best score (across scorers) · % ↗
1 more reported result
33% · Grok 4.6 (high) · Chess Puzzles · Best score (across scorers) · % · Epoch AI · Chess Puzzles ↗
Grok 4.6 (xhigh) · EBR-bench · Best score (across scorers) · % ↗
Grok 4.6 (xhigh) · Mystery Game Puzzles · Best score (across scorers) · % ↗
ARC-AGI-1 ↗87.5%
Grok 4.6 (medium) · ARC-AGI · Score · % ↗
3 more reported results
87% · Grok 4.6 (xhigh) · ARC-AGI · Score · % · Epoch AI · ARC-AGI ↗
87% · Grok 4.6 (high) · ARC-AGI · Score · % · Epoch AI · ARC-AGI ↗
74.83% · Grok 4.6 (low) · ARC-AGI · Score · % · Epoch AI · ARC-AGI ↗
WeirdML (v2) ↗67.29%
Grok 4.6 (high) · WeirdML · Accuracy · % ↗
ARC-AGI-2 ↗67.08%
Grok 4.6 (xhigh) · ARC-AGI-2 · Score · % ↗
3 more reported results
65.14% · Grok 4.6 (high) · ARC-AGI-2 · Score · % · Epoch AI · ARC-AGI-2 ↗
61.25% · Grok 4.6 (medium) · ARC-AGI-2 · Score · % · Epoch AI · ARC-AGI-2 ↗
27.64% · Grok 4.6 (low) · ARC-AGI-2 · Score · % · Epoch AI · ARC-AGI-2 ↗
CursorBench 4.0 ↗41.4%
Grok 4.6 (xhigh) · CursorBench · Score · % ↗
3 more reported results
40.4% · Grok 4.6 (high) · CursorBench · Score · % · Epoch AI · CursorBench ↗
36.1% · Grok 4.6 (medium) · CursorBench · Score · % · Epoch AI · CursorBench ↗
33.4% · Grok 4.6 (low) · CursorBench · Score · % · Epoch AI · CursorBench ↗
CritPt ↗19.71%
Grok 4.6 (xhigh) · CritPt · Accuracy · % ↗
3 more reported results
17.71% · Grok 4.6 (medium) · CritPt · Accuracy · % · Epoch AI · CritPt ↗
17.14% · Grok 4.6 (high) · CritPt · Accuracy · % · Epoch AI · CritPt ↗
5.71% · Grok 4.6 (low) · CritPt · Accuracy · % · Epoch AI · CritPt ↗
DeepSWE v1.1 ↗67.48%
Grok 4.6 (medium) · DeepSWE · Pass@1 · % ↗
3 more reported results
66.74% · Grok 4.6 (xhigh) · DeepSWE · Pass@1 · % · Epoch AI · DeepSWE ↗
65.19% · Grok 4.6 (high) · DeepSWE · Pass@1 · % · Epoch AI · DeepSWE ↗
41.65% · Grok 4.6 (low) · DeepSWE · Pass@1 · % · Epoch AI · DeepSWE ↗
LMCA ↗48.45%
Grok 4.6 (xhigh) · LMCA · Score · % ↗
DTBench ↗97.33%
Grok 4.6 (xhigh) · DTBench · Accuracy · % ↗
Grok 4.6 (xhigh) · Furniture Assembly · Best score (across scorers) · % ↗
GDP.pdf ↗17.2%
Grok 4.6 (xhigh) · GDP.pdf · GDP.pdf score · % ↗
1 more reported result
16% · Grok 4.6 (high) · GDP.pdf · GDP.pdf score · % · Epoch AI · GDP.pdf ↗
FrontierSWE v2 ↗25.29%
Grok 4.6 (xhigh) · FrontierSWE · Score · % ↗
SciCode ↗56.48%
Grok 4.6 (high) · SciCode · Score · % ↗
3 more reported results
55.9% · Grok 4.6 (medium) · SciCode · Score · % · Epoch AI · SciCode ↗
53.01% · Grok 4.6 (xhigh) · SciCode · Score · % · Epoch AI · SciCode ↗
49.42% · Grok 4.6 (low) · SciCode · Score · % · Epoch AI · SciCode ↗
Text Arena (Coding) ↗1617.64 Elo
Grok 4.6 (high) · WebDev Arena · Elo ↗
Epoch Capabilities Index (ECI) ↗156.57 ECI
Grok 4.6 · General ECI · 26 October 2026 fit ↗
Vals RSI Index ↗0.38 reference score
Grok 4.6 (xhigh) · Compression · reference-anchored score ↗
3 more reported results
0.42 reference score · Grok 4.6 (xhigh) · LM training · reference-anchored score · Vals RSI Index · LM training ↗
0.22 reference score · Grok 4.6 (xhigh) · Harness engineering · reference-anchored score · Vals RSI Index · Harness engineering ↗
0.00 reference score · Grok 4.6 (xhigh) · Post-training · reference-anchored score · Vals RSI Index · Post-training ↗
Vals Science Index ↗38.09 %
Grok 4.6 · Vals Science Index · Accuracy ↗
Legal Research Bench ↗48.08 %
Grok 4.6 · Legal Research Bench · All-Pass ↗
LegalBench ↗86.31 %
Grok 4.6 · LegalBench · Accuracy ↗
Grok 4.6 · Harvey's Legal Agent Benchmark v1 · Accuracy ↗
Tax Agent Bench ↗33.14 %
Grok 4.6 · Tax Agent Bench · Accuracy ↗
Vals Finance Agent v2 ↗53.68 %
Grok 4.6 · Finance Agent v2 · Accuracy ↗
Grok 4.6 · Excel Modeling Benchmark v1 · Accuracy ↗
CorpFin v2 ↗66.16 %
Grok 4.6 · CorpFin · Accuracy ↗
MortgageTax & TaxEval v2 ↗64.19 %
Grok 4.6 · MortgageTax · Accuracy ↗
1 more reported result
71.1 % · Grok 4.6 · TaxEval · Accuracy · Vals AI · TaxEval ↗
MedCode ↗44.71 %
Grok 4.6 · MedCode · Accuracy ↗
MedScribe ↗86.53 %
Grok 4.6 · MedScribe · Accuracy ↗
ProofBench ↗51 %
Grok 4.6 · ProofBench · Accuracy ↗
Terminal-Bench-Science 0.1 ↗11.43 %
Grok 4.6 (Mini-SWE-agent) · Terminal-Bench-Science · Score ↗
BioMysteryBench ↗72.22 %
Grok 4.6 · BioMysteryBench · Accuracy ↗
MysteryMechanism ↗30.63 %
Grok 4.6 · MysteryMechanism · Accuracy ↗
MMLU-Pro ↗89.4 %
Grok 4.6 · MMLU Pro · Accuracy ↗
Grok 4.6 · SAGE · Accuracy ↗
Vibe Code Bench ↗14.75 %
Grok 4.6 · Vibe Code Bench · Vibe Code Bench 1–100 ↗
1 more reported result
76.24 % · Grok 4.6 (OpenHands) · Vibe Code Bench · original task set · Vals AI · Vibe Code Bench ↗
Grok 4.6 · ProgramBench · Accuracy ↗
Terminal-Bench 4.0 ↗17.17 %
Grok 4.6 (Mini-SWE-agent) · Terminal-Bench 4.0 · Score ↗
SkillsBench ↗55.77 %
Grok 4.6 (OpenHands) · SkillsBench · Score ↗
IOI v1 & IOI v2 ↗47.61 %
Grok 4.6 · IOI · Accuracy ↗
Code Migration v1.1 ↗44.57 %
Grok 4.6 · Code Migration v1 · Accuracy ↗
SWE-bench Verified ↗95.6 %
Grok 4.6 (Mini-SWE-agent) · SWE-bench Verified · Score ↗
LiveCodeBench ↗88.22 %
Grok 4.6 · LiveCodeBench · Accuracy ↗
CyberBench v1.1 ↗67.68 %
Grok 4.6 · CyberBench · Accuracy ↗
Grok 4.6 · Time Horizon Index: KSP · Overall ↗
Public Benefits Bench v1.1 ↗66.85 %
Grok 4.6 · Public Benefits Bench · Accuracy ↗
Vending-Bench 2 ↗9047.0325 USD
Grok 4.6 · End-of-year balance ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
Grok 4.7 →xAI
24 of 39 shared benchmark comparisons
Claude Opus 4.8 →Anthropic
27 of 39 shared benchmark comparisons
Gemini 3.8 Flash →Google
32 of 39 shared benchmark comparisons
How similarity is measured
Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
- ARC Prize - Leaderboard ↗
- Arena leaderboard — WebDev evaluation ↗
- AA-AnalystAgent Benchmark | Artificial Analysis ↗
- EnterpriseOps-Gym-AA Benchmark | Artificial Analysis ↗
- GPQA Diamond Benchmark | Artificial Analysis ↗
- SciCode Benchmark | Artificial Analysis ↗
- 𝜏³-Banking Benchmark | Artificial Analysis ↗
- Terminal-Bench 2.1 Benchmark | Artificial Analysis ↗
- DTBench — Conceptual Reasoning Index ↗
- LMCA — Conceptual Reasoning Index ↗
- Cursor · CursorBench ↗
- DeepSWE ↗
- ARC-AGI-2 ↗
- Chess Puzzles ↗
- CritPt v1 ↗
- Earthborne Rangers (EBR-bench) ↗
- FrontierMath Tier 4 (v2) ↗
- FrontierMath Tiers 1-3 (v2) ↗
- Furniture Assembly Benchmark ↗
- GPQA Diamond ↗
- Mystery Game Puzzles ↗
- OTIS Mock AIME 2024-2025 ↗
- SimpleQA Verified ↗
- Vending Bench 2 — methodology and published results ↗
- AI benchmarks & capabilities ↗
- WeirdML ↗
- GDP.pdf Benchmark | Surge AI ↗
- FrontierSWE — software engineering benchmark ↗
- BioMysteryBench — evaluation methodology and results ↗
- Code Migration v1 — evaluation methodology and results ↗
- CorpFin — evaluation methodology and results ↗
- CyberBench — evaluation methodology and results ↗
- Excel Modeling Benchmark v1 — evaluation methodology and results ↗
- Finance Agent v2 — evaluation methodology and results ↗
- GPQA Diamond — evaluation methodology and results ↗
- Harvey's Legal Agent Benchmark v1 — evaluation methodology and results ↗
- IOI — evaluation methodology and results ↗
- LiveCodeBench — evaluation methodology and results ↗
- LegalBench — evaluation methodology and results ↗
- Legal Research Bench — evaluation methodology and results ↗
- MedCode — evaluation methodology and results ↗
- MedScribe — evaluation methodology and results ↗
- MMLU Pro — evaluation methodology and results ↗
- MortgageTax — evaluation methodology and results ↗
- MysteryMechanism — evaluation methodology and results ↗
- ProgramBench — evaluation methodology and results ↗
- ProofBench — evaluation methodology and results ↗
- Public Benefits Bench — evaluation methodology and results ↗
- Vals RSI Index — evaluation methodology and results ↗
- SAGE — evaluation methodology and results ↗
- Vals Science Index — evaluation methodology and results ↗
- SkillsBench — evaluation methodology and results ↗
- SWE-bench Verified — evaluation methodology and results ↗
- Tax Agent Bench — evaluation methodology and results ↗
- TaxEval — evaluation methodology and results ↗
- Terminal-Bench 4.0 — evaluation methodology and results ↗
- Terminal-Bench-Science — evaluation methodology and results ↗
- Time Horizon Index: KSP — evaluation methodology and results ↗
- Vibe Code Bench 1-100 — evaluation methodology and results ↗
- Vibe Code Bench — evaluation methodology and results ↗
Names used in source tables
Grok 4.6 · Grok 4.6 (High) · Grok 4.6 (Mini-SWE-agent) · Grok 4.6 (OpenHands) · Grok 4.6 (high) · Grok 4.6 (low) · Grok 4.6 (medium) · Grok 4.6 (xhigh)