Benchmark

xAI — reported benchmarks

xAI · source page ↗ · last checked Oct 4, 2026, 12:04 PM

Reported benchmarks · Grok 4.7

captured Oct 4, 2026, 12:04 PM
BenchmarkScore
CursorBench 4.046.3% · xhigh effort
DeepSWE v1.171.0% · high-effort score
EEBench64.0% · electrical engineering
AA Briefcase v1.11657 Elo · xhigh effort
Terminal-Bench 4.037.6% · agentic coding, xhigh effort
Harvey Legal Agent Benchmark19.6%
HealthBench Professional56.7%
GDPval1695 Elo · xhigh effort
LatchBio Biosafety Benchmark62.4% · dual-use domains
HackerBench v0.33.3% · risky dual-use prompts allowed through

Vendor-reported via automated web search — not independently verified. See the cited matrix on /models.

In the news · xAI

Importance-filtered press coverage (Google News) mentioning xAI. Headlines link to the original; verify before acting.

Change history