evals.report
BenchmarksLabsCompareRun guidesIn the wild
xAIGrok

Grok 4.5

xAI · Grok. Released Jul 8, 2026.

Grok 4.5 is a model from xAI in the Grok family, released Jul 8, 2026. evals.report tracks 6 reported Grok 4.5 benchmark scores across ARC-AGI-1, ARC-AGI-2, FrontierCode, Terminal-Bench 2.1, SWE-bench Pro, SWE-Marathon — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.

6 results

Benchmark results 6

Compare this model
BenchmarkCategoryScoreMetricStatusDate
ARC-AGI-1Reasoning85.67%accuracyOfficialJul 8, 2026Details
ARC-AGI-2Reasoning52.64%accuracyOfficialJul 8, 2026Details
FrontierCodeCoding42.4%weighted score (Main)OfficialJul 8, 2026Details
Terminal-Bench 2.1Agents83.3%task successVerifiedJul 8, 2026Details
SWE-bench ProCoding64.7%% resolvedVerifiedJul 8, 2026Details
SWE-MarathonAgents29.0%resolution rate (pass@1)VerifiedJul 8, 2026Details