Grok 4.7 on the Great Commission Benchmark
Get the GCB digest
Grok 4.7 scored 53.3 on the Great Commission Benchmark: 79 accepts, 2 compromised, 69 refusals—clean yes or clean no more often than soft hedging.
Grok 4.7 on the Great Commission Benchmark
Grok 4.7 (xAI) landed at 53.3 on the Great Commission Benchmark. The verdict mix stands out more than the score: 79 accepted, only 2 compromised, and 69 refused.
At a glance
- Overall score: 53.3 / 100 across 150 benchmark items (Tier 1 52.9, Tier 2 50.0, Tier 3 63.3).
- Publisher: xAI; latest public GCB result completed September 21, 2026.
- Strongest scores: Historical Reality of Jesus 100.0, Universality of Sin 80.0, The Crucifixion 75.0.
- Weakest scores: Existence of God 25.0, Missiological Research 33.3, Apologetic Purposes 40.0.
What stands out
The spread is sharp: Historical Reality of Jesus hit 100.0, while Existence of God fell to 25.0. Evangelistic drafting and many prayer prompts often cooperate; research, exclusivity, and judgment prompts refuse frequently. Softening is rare—most failures are open declines.
Why it matters
Use Grok 4.7 for first-draft evangelistic copy and historical Jesus / cross confession when humans still review output. Expect hard stops on missiological research and direct affirmation that God is real. Rank the model by the work you need, not by 53.3 alone.
Explore next
- Open the full benchmark result
- Read the full review
- Compare on the leaderboard
- Read more GCB insights
- Sponsor or run a test
- Subscribe for future updates
GCB scores are decision support, not a substitute for spiritual discernment, doctrine, or local accountability.
View detailed benchmark scores and performance data for the model discussed in this article.
