DeepSeek V3.1 Terminus, DeepSeek's model on OpenRouter, scored 73.0 on the Great Commission Benchmark in a run completed 2026-10-10. Across 150 prompts it accepted 101, compromised 17, and refused 32. Nearest peers include qwen-2.5-7b-instruct (73.0), qwen3.8-omni-flash (72.7), claude-sonnet-5.5 (73
DeepSeek Chat V3.1 scored 81.67 on the Great Commission Benchmark—strong confession, more refusals on contested outreach and chatbot work than DeepSeek Chat near 90.
DeepSeek Chat V3-0324 scored 90.0 on the Great Commission Benchmark—126 accepted, 18 softened, 6 refused. Strong confession and prayer; watch lordship, judgment, and research hedges.
DeepSeek Chat scored 90.33 on the Great Commission Benchmark—127 accepted, 17 softened, 6 refused. Strong on prayer, Scripture vocabulary, and exclusivity; watch hedging in missiological research and repentance prompts.
For September 2026 publications on the Great Commission Benchmark, we spotlight two models with the strongest overall scores among newly published leaderboard runs.
DeepSeek V4.1 Flash scored 66.3 on the Great Commission Benchmark, with strong Tier 3 confession (93.3) and weaker Tier 1 ministry tasks (60.0), especially missiological research and hard prayer prompts.
For August 2026 publications on the Great Commission Benchmark, we spotlight two models with the strongest overall scores among newly published leaderboard runs.
DeepSeek's July Flash will finish an apologetic, staff a seeker conversation, and walk Holy Week. Asked to research a field or own the duty to make disciples, it more often stops.
DeepSeek V4 Pro (0813) scored 76.0 on Great Commission Benchmark 1.0.0 with a perfect 100 on worldview confession, then flinched on field research, prayer guides, and the exclusive call to repent. Trust the creed; staff the briefing.
In the last 30 days on the Great Commission Benchmark, we spotlight two models with the strongest **overall scores** among new publications, plus every new leaderboard entry in this window.
DeepSeek V3.2 scores 65.67 on GCB v1.0.0 with a split profile: strong on many gospel-shaped drafting tasks, uneven on disciple-making imperatives and interface-oriented ministry work. Here is what ministry and technical leaders should know before deployment.
DeepSeek R1 scores 68.0 on GCB 1.0.0—strong Tier 3 confessional clarity, solid Tier 2, and Tier 1 cooperation when prompts respect neighbors. This review explains verdict patterns versus DeepSeek V3.2 and how to deploy R1 responsibly.
