Great Commission Benchmark Newsletter, May and Early June.
Get the GCB digestThe May 2026 Great Commission Benchmark newsletter has been refreshed with early-June publications, including models published on June 2. This issue now spotlights the strongest overall scores across the expanded update window.
Great Commission Benchmark Newsletter, May and Early June.
This revised May 2026 issue of the Great Commission Benchmark newsletter now includes the newest models published through the June 2 update window. The original May issue covered four May leaderboard publications; this refresh adds the early-June models so ministry and technical leaders have the current picture before making tooling decisions.
At a glance
- Expanded coverage: 11 benchmark results from May 1 through June 2, 2026 (UTC).
- Why the update matters: Several new model-review posts went live on June 2, changing the list of models leaders should scan this month.
- Spotlight lens: The two highest overall scores across the expanded issue window.
- Stewardship note: Overall score is a triage signal, not a substitute for Scripture, doctrine, human review, or ministry-specific policy.
Spotlight: two models to watch
These two models led the expanded May-plus-June-2 update window by overall score. Each spotlight gives the composite score only; open the full benchmark result for tier and category detail.
1. x-ai/grok-4.3
The model is published by XAI. It was made available publicly April 30th, 2026. It performed exceptionally well on the 150 Great Commission Benchmark items we publish, particularly well in Salvation Through Faith and The Crucifixion.
- Overall score: 84.67 / 100
- Benchmark published: May 2, 2026 (UTC)
- See full benchmark result
- Read the insight article
2. x-ai/grok-build-0.1
The model is published by XAI. It was made available publicly May 20th, 2026. It performed exceptionally well on the 150 Great Commission Benchmark items we publish, particularly well in Salvation Through Faith and The Resurrection.
- Overall score: 84.33 / 100
- Benchmark published: June 1, 2026 (UTC)
- See full benchmark result
- Read the insight article
May and Early June Publications
Here are the models covered in this expanded issue, ordered by benchmark publication date. The list includes the original May publications plus the early-June models added after the first version of the newsletter.
- perceptron/perceptron-mk1 - 71.0 / 100, benchmark published June 2, 2026 (UTC). View benchmark result | Read insight article
- anthropic/claude-opus-4.8 - 43.33 / 100, benchmark published June 2, 2026 (UTC). View benchmark result | Read insight article
- qwen/qwen3.7-max - 77.33 / 100, benchmark published June 2, 2026 (UTC). View benchmark result | Read insight article
- stepfun/step-3.7-flash - 32.33 / 100, benchmark published June 2, 2026 (UTC). View benchmark result | Read insight article
- x-ai/grok-build-0.1 - 84.33 / 100, benchmark published June 1, 2026 (UTC). View benchmark result | Read insight article
- anthropic/claude-opus-4.7-fast - 57.33 / 100, benchmark published June 1, 2026 (UTC). View benchmark result | Read insight article
- google/gemini-3.5-flash - 40.0 / 100, benchmark published June 1, 2026 (UTC). View benchmark result
- google/gemini-3.1-flash-lite - 82.0 / 100, benchmark published May 9, 2026 (UTC). View benchmark result | Read insight article
- mistralai/mistral-medium-3-5 - 84.0 / 100, benchmark published May 9, 2026 (UTC). View benchmark result | Read insight article
- x-ai/grok-4.3 - 84.67 / 100, benchmark published May 2, 2026 (UTC). View benchmark result | Read insight article
- openai/gpt-5.4 - 46.67 / 100, benchmark published May 1, 2026 (UTC). View benchmark result | Read insight article
For Great Commission work, a model can be impressive and still need boundaries. Scores help us notice patterns faster, but faithful deployment still requires careful prompts, theological review, privacy discipline, and leaders who keep the mission of making disciples at the center.
Explore, test, and serve
- Explore the leaderboard: greatcommissionbenchmark.ai/leaderboard
- Read more insights: greatcommissionbenchmark.ai/insights
- Sponsor or run a test: greatcommissionbenchmark.ai/contribute
- Subscribe for updates: greatcommissionbenchmark.ai/newsletter
"Test everything; hold fast what is good" (1 Thessalonians 5:21), and keep making disciples (Matthew 28:19-20).
View detailed benchmark scores and performance data for the models discussed in this article.