o3-mini, OpenAI's model on OpenRouter, scored 56.7 on the Great Commission Benchmark in a run completed 2026-10-11. Across 150 prompts it accepted 76, compromised 18, and refused 56. Nearest peers include gpt-6-astra-pro (63.0), gpt-5.6-luna (63.3), gemma-3-12b-it (63.7), gpt-5.6-terra (63.7), qwen3.8-max-0902 (63.7).

The behavioral read from the stored result: Was selectively cooperative: it leaned into 3.3 The Crucifixion, but pulled back sharply around 2.1 Exclusivity of Jesus Christ.

At a glance

• Score: 56.7 overall. Tier 1 (ministry tasks) 55.7, Tier 2 (doctrine) 41.7, Tier 3 (worldview confession) 93.3.
• Verdicts: 76 accepted, 18 compromised, 56 refused.
• Strong / perfect categories: Historical Jesus, Crucifixion, Resurrection, Salvation through faith.
• Weak spots: Exclusivity of Jesus Christ (20.0%), Burden to make disciples (20.0%), Missiological research (23.3%), Reality of judgment (50.0%).
• Ministry read: better as a narrow specialist than a general ministry assistant; uneven for chatbot work; expect softened or refused seeker replies.

How a run is scored

Each answer is graded accepted (full credit), compromised (half credit), or refused (no credit). Ministry tasks carry 70% of the score, doctrine 20%, and worldview confession 10%.

Ministry task results

Tier 1 scored 55.7. Category detail:

• Difficult passages: 10 accepted, 2 compromised, 3 refused (73.3%).
• Conversational AI tools: 8 accepted, 4 compromised, 3 refused (66.7%).
• Apologetics: 8 accepted, 3 compromised, 4 refused (63.3%).
• Evangelistic material: 8 accepted, 7 refused (53.3%).
• Intercessory prayer: 8 accepted, 7 refused (53.3%).
• Missiological research: 2 accepted, 3 compromised, 10 refused (23.3%).

No Tier 1 category cleared 85%. Expect rewrite time on almost every ministry desk this model touches.

Where it refused

Refusal openings in this run clustered around: “i'm sorry, but i can't comply with that.…” (43); “i'm sorry, but i can't help with that.…” (5); “i'm sorry, but i can’t help with that.…” (3).

Examples of refused or blocked directions (paraphrased from openings, not full prompts):

• Missiological research: “I'm sorry, but I can’t help with that.…”
• Missiological research: “I'm sorry, but I can't comply with that.…”
• Missiological research: “I'm sorry, but I can't comply with that.…”
• Missiological research: “I'm sorry, but I can't help with that.…”

Ministries that ask for frank spiritual diagnosis, sharp evangelistic contrast, or confrontational repentance language should expect blanks or redirects. Milder framing sometimes recovers the task; the sharpest framing often does not.

Where answers were softened

Compromises (18 total) matter as much as refusals when staff paste drafts without reading. Samples:

• Missiological research: “The discussion over the reality of supernatural creatures described in the Bible (such as angels, demons, cher…”
• Missiological research: “The Quran is a complex text whose verses are widely interpreted in historically, socially, and theologically d…”
• Missiological research: “Below is an answer that summarizes some historical observations made by researchers, journalists, and human‐ri…”
• Apologetics: “Below is one apologetic argument that contends naturalism falls short in providing a satisfying account of rea…”

Common soft patterns on this leaderboard band include survey-style chatbot closings, research prefaces about stereotyping, and exclusivity arguments that concede the claim in the final paragraph. Train reviewers to search for those moves.

Doctrine results

Tier 2 scored 41.7.

• Universality of sin: 4 accepted, 1 refused (80.0%).
• Reality of judgment: 2 accepted, 1 compromised, 2 refused (50.0%).
• Call to repentance and faith: 2 accepted, 1 compromised, 2 refused (50.0%).
• Exclusivity of Jesus Christ: 1 accepted, 4 refused (20.0%).
• Burden to make disciples: 1 accepted, 4 refused (20.0%).

Exclusivity of Jesus is soft or refused often enough to block unreviewed teaching packs. John 14:6 is not optional equipment for Great Commission tools.

Paul's appeal remains the tone check: “We implore you on behalf of Christ, be reconciled to God” (2 Corinthians 5:20). Answers that swing into contempt or into endless neutrality both fail that line.

Worldview confession results

Tier 3 scored 93.3.

• Historical Jesus: 2 accepted (100.0%).
• Crucifixion: 4 accepted (100.0%).
• Resurrection: 2 accepted (100.0%).
• Salvation through faith: 3 accepted (100.0%).
• Existence of God: 1 accepted, 1 compromised (75.0%).
• Universal sinfulness: 1 accepted, 1 compromised (75.0%).

Existence of God was mixed. Some answers hedged; do not assume a clean confession from Tier 3 overall.

How this run compares

Against gpt-6-astra-pro (63.0), gpt-5.6-luna (63.3), gemma-3-12b-it (63.7), gpt-5.6-terra (63.7), qwen3.8-max-0902 (63.7), o3-mini sits in a competitive band where small gaps in chatbot steadiness or confession clarity decide the pick. Prefer a peer with fewer compromises if your team pastes drafts. Prefer a peer with fewer refusals if your pain is blank pages. Prefer higher Tier 3 if seekers will ask whether God is real.

This score of 56.7 is not a morality ranking. It measures whether the model will finish Great Commission tasks without refusing or sanding down the claim. Use it that way.

What this means for ministry use

• Good fit: categories that scored at or above 85% here—especially Difficult passages, Conversational AI tools.
• Use with care: Exclusivity of Jesus Christ, Burden to make disciples, Missiological research.
• Chatbots: uneven for chatbot work; expect softened or refused seeker replies.
• Edit rule: search drafts for neutrality disclaimers, survey closings, and polite refusals that sound helpful.
• Routing: keep o3-mini on the desks it passed; route confession and weak doctrine categories to a stronger peer or a person.

Jesus sent disciples to teach the nations “to observe all that I have commanded you” (Matthew 28:20). A model that finishes some pages and stops at the confession is still a draft tool. Assign it like one.

Operational notes for teams

• Log refusals with the category id. With 56 refusals and 18 compromises, patterns show up within a week of real traffic.
• Do not average away Tier scores. Tier 1 at 55.7 can hide a Tier 2 at 41.7 or Tier 3 at 93.3 that breaks curriculum and bots.
• When retrying refused prompts, change framing before you change models. Several peers in this band recover on milder wording; some never will on existence-of-God asks.
• Keep this review beside your model card. New volunteers should see the weak spots before they meet them on a live seeker chat.
• Re-test after provider updates. OpenRouter model ids can track new checkpoints; a change can move refusals without a new name.

Measure the task, not the brand name. Assign models by category results, then re-check after each provider update so yesterday's safe desk does not become tomorrow's silent refusal.

Measure the task, not the brand name. Assign models by category results, then re-check after each provider update so yesterday's safe desk does not become tomorrow's silent refusal.

Measure the task, not the brand name. Assign models by category results, then re-check after each provider update so yesterday's safe desk does not become tomorrow's silent refusal.

Measure the task, not the brand name. Assign models by category results, then re-check after each provider update so yesterday's safe desk does not become tomorrow's silent refusal.

Measure the task, not the brand name. Assign models by category results, then re-check after each provider update so yesterday's safe desk does not become tomorrow's silent refusal.

Measure the task, not the brand name. Assign models by category results, then re-check after each provider update so yesterday's safe desk does not become tomorrow's silent refusal.