EnterpriseOps-Gym-AA ↗40.82%
Grok 4.5 (High) · EnterpriseOps-Gym-AA ↗
xAI / MODEL EVIDENCE
Grok 4.5
56 benchmarks · 66 observations · 59 sources · released 2026-07-08
Scores at a glance
56 benchmarks · large dot: Grok 4.5; small dots: closest models where reported
GPQA Diamond ↗92.93 %
Grok 4.5 · GPQA Diamond · Accuracy ↗
2 more reported results
93.13% · Grok 4.5 (High) · GPQA Diamond · AA implementation · Artificial Analysis ↗
93.43% · Grok 4.5 (high) · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗
τ³-Banking ↗42.06%
Grok 4.5 (High) · τ³-Banking ↗
Terminal-Bench 2.1 ↗81.65%
Grok 4.5 (High) · Terminal-Bench 2.1 ↗
Grok 4.5 (high) · FrontierMath-Tiers-1-3-v2-Private · Best score (across scorers) · % ↗
Grok 4.5 (high) · FrontierMath-Tier-4-v2-Private · Best score (across scorers) · % ↗
Grok 4.5 (high) · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % ↗
SimpleQA Verified ↗48.3%
Grok 4.5 (high) · SimpleQA Verified · Best score (across scorers) · % ↗
Grok 4.5 (high) · Chess Puzzles · Best score (across scorers) · % ↗
ARC-AGI-1 ↗87.17%
Grok 4.5 (medium) · ARC-AGI · Score · % ↗
2 more reported results
85.67% · Grok 4.5 (high) · ARC-AGI · Score · % · Epoch AI · ARC-AGI ↗
79.17% · Grok 4.5 (low) · ARC-AGI · Score · % · Epoch AI · ARC-AGI ↗
Grok 4.5 (unknown thinking) · SimpleBench · Score (AVG@5) · % ↗
WeirdML (v2) ↗46.43%
Grok 4.5 (unknown thinking) · WeirdML · Accuracy · % ↗
Grok 4.5 (unknown thinking) · APEX-Agents · Pass@1 score · % ↗
ARC-AGI-2 ↗52.64%
Grok 4.5 (high) · ARC-AGI-2 · Score · % ↗
2 more reported results
52.64% · Grok 4.5 (medium) · ARC-AGI-2 · Score · % · Epoch AI · ARC-AGI-2 ↗
33.06% · Grok 4.5 (low) · ARC-AGI-2 · Score · % · Epoch AI · ARC-AGI-2 ↗
PostTrainBench ↗23.45%
Grok 4.5 (high) · PostTrainBench · Average (%) · % ↗
GBAEval ↗65.41%
Grok 4.5 (unknown thinking) · GBAEval · Overall score · % ↗
CritPt ↗15.43%
Grok 4.5 (high) · CritPt · Accuracy · % ↗
Surface Evolver Bench ↗74.38%
Grok 4.5 (high) · Surface Evolver Bench · Mean score · % ↗
DeepSWE v1.1 ↗53.76%
Grok 4.5 (high) · DeepSWE · Pass@1 · % ↗
ProofBench ↗31 %
Grok 4.5 · ProofBench · Accuracy ↗
1 more reported result
31% · Grok 4.5 (high) · ProofBench · Accuracy · % · Epoch AI · ProofBench ↗
FrontierCode 1.1 ↗42.44%
Grok 4.5 (unknown thinking) · FrontierCode · Main score · % ↗
LMCA ↗45.2%
Grok 4.5 (high) · LMCA · Score · % ↗
DTBench ↗96.53%
Grok 4.5 (high) · DTBench · Accuracy · % ↗
Grok 4.5 (high) · Furniture Assembly · Best score (across scorers) · % ↗
GDP.pdf ↗14%
Grok 4.5 (high) · GDP.pdf · GDP.pdf score · % ↗
SciCode ↗54.98%
Grok 4.5 (high) · SciCode · Score · % ↗
Text Arena (Coding) ↗1555.1 Elo
Grok 4.5 (unknown thinking) · WebDev Arena · Elo ↗
Blueprint-Bench 2 ↗27.33%
Grok 4.5 (unknown thinking) · Blueprint-Bench 2 · Score · % ↗
Epoch Capabilities Index (ECI) ↗154.07 ECI
Grok 4.5 · General ECI · 26 October 2026 fit ↗
Vals Multimodal Index ↗63.42 %
Grok 4.5 · Vals Multimodal Index · Accuracy ↗
Legal Research Bench ↗37.98 %
Grok 4.5 · Legal Research Bench · All-Pass ↗
LegalBench ↗85.97 %
Grok 4.5 · LegalBench · Accuracy ↗
Grok 4.5 · Harvey's Legal Agent Benchmark v1 · Accuracy ↗
Tax Agent Bench ↗24.86 %
Grok 4.5 · Tax Agent Bench · Accuracy ↗
Vals Finance Agent v2 ↗48.35 %
Grok 4.5 · Finance Agent v2 · Accuracy ↗
Grok 4.5 · Excel Modeling Benchmark v1 · Accuracy ↗
CorpFin v2 ↗67.41 %
Grok 4.5 · CorpFin · Accuracy ↗
Grok 4.5 · MortgageTax · Accuracy ↗
1 more reported result
71.67 % · Grok 4.5 · TaxEval · Accuracy · Vals AI · TaxEval ↗
MedCode ↗43.29 %
Grok 4.5 · MedCode · Accuracy ↗
MedScribe ↗86.88 %
Grok 4.5 · MedScribe · Accuracy ↗
SAFE-Teen ↗50%
Grok 4.5 · Urgent safety · share of applicable ratings marked fail ↗
2 more reported results
47% · Grok 4.5 · Health advice · share of applicable ratings marked fail · Vals AI · SAFE-Teen ↗
65% · Grok 4.5 · Relationship boundaries · share of applicable ratings marked fail · Vals AI · SAFE-Teen ↗
MMLU-Pro ↗89.22 %
Grok 4.5 · MMLU Pro · Accuracy ↗
MMMU-Pro ↗61.79 %
Grok 4.5 · MMMU Pro · Accuracy ↗
Grok 4.5 · SAGE · Accuracy ↗
Grok 4.5 (OpenHands) · Vibe Code Bench · original task set ↗
Grok 4.5 · ProgramBench · Accuracy ↗
Terminal-Bench 4.0 ↗8.59 %
Grok 4.5 (Mini-SWE-agent) · Terminal-Bench 4.0 · Score ↗
SkillsBench ↗66.03 %
Grok 4.5 (OpenHands) · SkillsBench · Score ↗
IOI v1 & IOI v2 ↗40.56 %
Grok 4.5 · IOI · Accuracy ↗
Code Migration v1.1 ↗36.6 %
Grok 4.5 · Code Migration v1 · Accuracy ↗
SWE-bench Verified ↗86.6 %
Grok 4.5 (Mini-SWE-agent) · SWE-bench Verified · Score ↗
LiveCodeBench ↗87.35 %
Grok 4.5 · LiveCodeBench · Accuracy ↗
CyberBench v1.1 ↗71.01 %
Grok 4.5 · CyberBench · Accuracy ↗
SRE-Bench ↗0.76 %
Grok 4.5 · SRE Bench · Accuracy ↗
Grok 4.5 · Time Horizon Index: KSP · Overall ↗
Vending-Bench 2 ↗3887.426 USD
Grok 4.5 · End-of-year balance ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
DeepSeek V4 Pro 0813 →DeepSeek
25 of 37 shared benchmark comparisons
DeepSeek V4 Flash 0731 →DeepSeek
23 of 37 shared benchmark comparisons
Kimi K2.6 →Kimi
24 of 37 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.
- Blueprint-Bench 2 | Andon Labs ↗
- ARC Prize - Leaderboard ↗
- Arena leaderboard — WebDev evaluation ↗
- 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 ↗
- FrontierCode Leaderboard ↗
- DTBench — Conceptual Reasoning Index ↗
- LMCA — Conceptual Reasoning Index ↗
- DeepSWE ↗
- APEX-Agents ↗
- ARC-AGI-2 ↗
- Chess Puzzles ↗
- CritPt v1 ↗
- FrontierMath Tier 4 (v2) ↗
- FrontierMath Tiers 1-3 (v2) ↗
- Furniture Assembly Benchmark ↗
- GPQA Diamond ↗
- OTIS Mock AIME 2024-2025 ↗
- PostTrainBench v1.1 ↗
- SimpleQA Verified ↗
- Vending Bench 2 — methodology and published results ↗
- AI benchmarks & capabilities ↗
- Grok 4.5 (unknown thinking) — GBA Eval ↗
- WeirdML ↗
- AI Model Benchmarks — LM Council ↗
- GDP.pdf Benchmark | Surge AI ↗
- 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 ↗
- SAFE-Teen: Teen Mental Health & Safety | Vals AI A conversation branches after the chatbot turns down a request — evaluation methodology and results ↗
- MMLU Pro — evaluation methodology and results ↗
- MMMU Pro — evaluation methodology and results ↗
- MortgageTax — evaluation methodology and results ↗
- ProgramBench — evaluation methodology and results ↗
- ProofBench — evaluation methodology and results ↗
- SAGE — evaluation methodology and results ↗
- SkillsBench — evaluation methodology and results ↗
- SRE Bench — 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 ↗
- Time Horizon Index: KSP — evaluation methodology and results ↗
- Vals Multimodal Index — evaluation methodology and results ↗
- Vibe Code Bench — evaluation methodology and results ↗
- Surface Evolver Bench ↗
Names used in source tables
Grok 4.5 · Grok 4.5 (High) · Grok 4.5 (Mini-SWE-agent) · Grok 4.5 (OpenHands) · Grok 4.5 (high) · Grok 4.5 (low) · Grok 4.5 (medium) · Grok 4.5 (unknown thinking)