Grok 4.5 — benchmark results

Provider: xAI. Access: API.

Unified ELO 1923 ± 26, rank #41 of 1719 rated models, from 30 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
SWE-Marathon29Pass@1 (%)100
Vals AI Harvey Legal Agent Bench12.92Accuracy (%)100
UGI Leaderboard62.32UGI Score99.1
Vals AI GPQA92.93Accuracy (%)97.4
Vals AI SWE-bench Verified86.6Resolved (%)96.9
Vals AI MedScribe86.88Accuracy (%)95.5
UGI - Natural Intelligence60.21NatInt Score95.4
Vals AI LiveCodeBench87.35Accuracy (%)95.1
AI Chess Leaderboard (Reasoning)1516Elo94.7
UGI - Writing59.96Writing Score93.9
SvelteBench98.9Average pass@1 (%)93.5
VoxelBench1796Rating93

Interactive version: aibenchmarks.dev/model?slug=grok-4-5 · How the rankings work · Data refreshed daily, snapshot 2026-07-09.