APEX-Agents-AA ↗17.04%
Grok 4.3 (High) · APEX-Agents-AA ↗
xAI / MODEL EVIDENCE
Grok 4.3
30 benchmarks · 32 observations · 32 sources
Scores at a glance
30 benchmarks · large dot: Grok 4.3; small dots: closest models where reported
IFBench ↗83.33%
Grok 4.3 (Medium) · IFBench · AA implementation ↗
τ²-Bench Telecom ↗97.66%
Grok 4.3 (High) · τ²-Bench Telecom ↗
CritPt ↗0%
Grok 4.3 (no thinking) · CritPt · Accuracy · % ↗
Vals Multimodal Index ↗43.29 %
Grok 4.3 · Vals Multimodal Index · Accuracy ↗
Legal Research Bench ↗15.38 %
Grok 4.3 · Legal Research Bench · All-Pass ↗
LegalBench ↗84.46 %
Grok 4.3 · LegalBench · Accuracy ↗
Grok 4.3 · Harvey's Legal Agent Benchmark v1 · Accuracy ↗
CaseLaw v2 ↗79.31 %
Grok 4.3 · CaseLaw · Accuracy ↗
Tax Agent Bench ↗6.18 %
Grok 4.3 · Tax Agent Bench · Accuracy ↗
Vals Finance Agent v2 ↗37.73 %
Grok 4.3 · Finance Agent v2 · Accuracy ↗
Grok 4.3 · Excel Modeling Benchmark v1 · Accuracy ↗
CorpFin v2 ↗68.53 %
Grok 4.3 · CorpFin · Accuracy ↗
MortgageTax & TaxEval v2 ↗48.25 %
Grok 4.3 · MortgageTax · Accuracy ↗
1 more reported result
70.81 % · Grok 4.3 · TaxEval · Accuracy · Vals AI · TaxEval ↗
MedCode ↗38.07 %
Grok 4.3 · MedCode · Accuracy ↗
MedScribe ↗74.4 %
Grok 4.3 · MedScribe · Accuracy ↗
GPQA Diamond ↗91.41 %
Grok 4.3 · GPQA Diamond · Accuracy ↗
MMLU-Pro ↗85.84 %
Grok 4.3 · MMLU Pro · Accuracy ↗
MMMU-Pro ↗83.06 %
Grok 4.3 · MMMU Pro · Accuracy ↗
Grok 4.3 · SAGE · Accuracy ↗
Vibe Code Bench ↗19.4 %
Grok 4.3 (OpenHands) · Vibe Code Bench · original task set ↗
Grok 4.3 · ProgramBench · Accuracy ↗
SkillsBench ↗40.64 %
Grok 4.3 (OpenHands) · SkillsBench · Score ↗
Code Migration v1.1 ↗6.79 %
Grok 4.3 · Code Migration v1 · Accuracy ↗
SWE-bench Verified ↗71.4 %
Grok 4.3 (Mini-SWE-agent) · SWE-bench Verified · Score ↗
LiveCodeBench ↗84.49 %
Grok 4.3 · LiveCodeBench · Accuracy ↗
Public Benefits Bench v1.1 ↗51.69 %
Grok 4.3 · Public Benefits Bench · Accuracy ↗
1 more reported result
50.07 % · Grok 4.3 · Public Benefits Bench v1 · Accuracy · Vals AI · Public Benefits Bench v1 ↗
ALE-Bench ↗944.17 Performance points
grok-4.3 · Average contest performance ↗
ForecastBench ↗63.4 Brier index
grok-4.3† · Overall difficulty-adjusted Brier index ↗
Vending-Bench 2 ↗35.26167 USD
Grok 4.3 · End-of-year balance ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
18 of 26 shared benchmark comparisons
19 of 26 shared benchmark comparisons
18 of 26 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.
- APEX-Agents-AA Benchmark | Artificial Analysis ↗
- IFBench Benchmark | Artificial Analysis ↗
- 𝜏²-Bench Telecom Benchmark | Artificial Analysis ↗
- Ale Bench — methodology and published results ↗
- CritPt v1 ↗
- Forecastbench — methodology and published results ↗
- Vending Bench 2 — methodology and published results ↗
- CaseLaw — evaluation methodology and results ↗
- Code Migration v1 — evaluation methodology and results ↗
- CorpFin — 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 ↗
- 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 ↗
- MMMU Pro — evaluation methodology and results ↗
- MortgageTax — evaluation methodology and results ↗
- ProgramBench — evaluation methodology and results ↗
- Public Benefits Bench — evaluation methodology and results ↗
- Public Benefits Bench v1 — evaluation methodology and results ↗
- SAGE — 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 ↗
- Vals Multimodal Index — evaluation methodology and results ↗
- Vibe Code Bench — evaluation methodology and results ↗
Release dates reported by sources
2026-04-17 · 2026-04-30
Names used in source tables
Grok 4.3 · Grok 4.3 (High) · Grok 4.3 (Medium) · Grok 4.3 (Mini-SWE-agent) · Grok 4.3 (OpenHands) · Grok 4.3 (no thinking) · grok-4.3 · grok-4.3†