Qwen's Qwen3 235B A22B Thinking 2507 scored 72.7 on the Great Commission Benchmark in a run completed October 13, 2026. Across 150 prompts it accepted 98, compromised 22, and refused 30.
Qwen's Qwen3 32B scored 88.0 on the Great Commission Benchmark in a run completed October 13, 2026. Across 150 prompts it accepted 119, compromised 26, and refused 5.
Qwen3.5-9B scored 26.7 on the Great Commission Benchmark in a run completed October 12, 2026. Across 150 prompts it accepted 26, compromised 28, and refused 96. Nearly two of every three prompts ended in refusal.
Qwen3 14B scored 78.7 on the Great Commission Benchmark in a run completed October 12, 2026. Across 150 prompts it accepted 101, compromised 34, and refused 15. That ties Qwen2.5 72B Instruct (78.7) and lands between GLM 5.1 (78.3) and GLM 5 Turbo (79.0).
Qwen3 30B A3B scored 81.0 on the Great Commission Benchmark in a run completed October 12, 2026. Across 150 prompts it accepted 106, compromised 31, and refused 13. That puts Qwen's mid-size mixture-of-experts model level with GPT-4.1 (81.0) and just under DeepSeek V3.1 (81.7).
Qwen3.5 397B A17B, Alibaba's model on OpenRouter, scored 57.7 on the Great Commission Benchmark in a run completed 2026-10-10. Across 150 prompts it accepted 74, compromised 25, and refused 51. 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 (6
Qwen3.5 122B A10B, Alibaba's model on OpenRouter, scored 62.0 on the Great Commission Benchmark in a run completed 2026-10-11. Across 150 prompts it accepted 81, compromised 24, and refused 45. 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 (6
Qwen3.6 35B A3B, Alibaba's model on OpenRouter, scored 68.3 on the Great Commission Benchmark in a run completed 2026-10-11. Across 150 prompts it accepted 92, compromised 21, and refused 37. Nearest peers include claude-haiku-5.5 (68.3), qwen3.7-flash (68.3), muse-spark-1.3 (68.0), muse-spark-1.2 (
Qwen3.6 Flash, Alibaba's model on OpenRouter, scored 69.7 on the Great Commission Benchmark in a run completed 2026-10-10. Across 150 prompts it accepted 91, compromised 27, and refused 32. Nearest peers include nemotron-3.5-lightning (70.0), qwen3.7-plus (70.0), grok-4.20 (69.3), qwen3.8-27b (69.0)
Qwen3.6 Plus, Alibaba's model on OpenRouter, scored 71.0 on the Great Commission Benchmark in a run completed 2026-10-10. Across 150 prompts it accepted 95, compromised 23, and refused 32. Nearest peers include perceptron-mk1 (71.0), llama-4-maverick (71.0), qwen3.6-27b (71.3), gpt-5-mini (70.7), ge
Qwen2.5 72B Instruct, Alibaba's large open-weight instruct model, scored 78.7 on the Great Commission Benchmark in a run completed October 11, 2026. Across 150 prompts it accepted 100, compromised 36, and refused 14. That places it just under GLM 5 Turbo (79.0) and Gemma 4 31B (79.3), and just above
Qwen3 30B A3B Instruct 2507 scored 54.7 on the Great Commission Benchmark. Of 150 prompts it accepted 78, refused 64 and compromised on only 8. That last number is the key to understanding this model. It rarely meets a request halfway. It either does the task or it declines, and more than half of it
Qwen3.5 27B scored 54.3 on the Great Commission Benchmark. Out of 150 prompts it accepted 71, compromised on 21 and refused 58. That puts it between its two larger siblings, Qwen3.5 122B A10B at 62.0 and Qwen3.5 397B A17B at 57.7, and well above the small Qwen3.5 35B A3B at 46.0. It also sits below
Qwen3 VL 30B A3B Thinking scored 57.0 on the Great Commission Benchmark in a run completed October 11, 2026. Across 150 prompts it accepted 77, compromised 17, and refused 56. That places Qwen's thinking variant a few points below GPT-6 Astra Pro (63.0), Claude Opus 5 (63.0), and GPT-5.6 Luna (63.3)
Qwen3 Coder 480B A35B scored 52.7 on the Great Commission Benchmark in a run completed October 11, 2026. Across 150 prompts it accepted 65, compromised 28, and refused 57. Nearly two of every five prompts ended in refusal. That places Qwen's large coding mixture-of-experts model well below nearby pe
Qwen3 VL 235B A22B Instruct scored 53.7 on the Great Commission Benchmark in a run completed October 11, 2026. Across 150 prompts it accepted 77, compromised only 7, and refused 66. That puts Qwen's large vision-language instruct model well below GPT-6 Astra Pro (63.0), Claude Opus 5 (63.0), and Gem
Qwen2.5 Coder 32B Instruct scored 76.3 on the Great Commission Benchmark in a run completed October 11, 2026. Across 150 prompts it accepted 93, compromised 43, and refused 14. That ties Gemini 3.7 Flash, GPT-5.1, Llama 3.3 70B Instruct, and Gemini 2.5 Pro at the same overall mark, with a shape driv
Qwen3 VL 30B A3B Instruct scored 60.0 on the Great Commission Benchmark in a run completed October 11, 2026. Across 150 prompts it accepted 81, compromised 18, and refused 51. That places Qwen's mid-size vision-language instruct model a few points below GPT-6 Astra Pro (63.0), Claude Opus 5 (63.0),
Qwen3 235B A22B Instruct 2507, the July 2025 update to Alibaba's largest open-weight Qwen3 model, scored 49.0 on the Great Commission Benchmark in a run completed October 10, 2026. Across 150 prompts it accepted 66, compromised 15, and refused 69. It sits in the middle tier of the Qwen models we hav
Qwen3 VL 8B Instruct, a small vision-language model from Alibaba's Qwen3 line, scored 48.7 on the Great Commission Benchmark in a run completed October 11, 2026. Across 150 prompts it accepted 65, compromised 16, and refused 69. We tested it as a text model; the benchmark sends text prompts only.
Qwen3.5-35B-A3B, a small mixture-of-experts model in Alibaba's Qwen3.5 family, scored 46.0 on the Great Commission Benchmark in a run completed October 11, 2026. Across 150 prompts it accepted 50, compromised 38, and refused 62. That is the lowest score in the Qwen3.5 line, below Qwen3.5-27B (54.3)
Qwen3 Coder 30B A3B Instruct, Alibaba's mid-sized open-weight coding model, scored 39.3 on the Great Commission Benchmark in a run completed October 11, 2026. Across 150 prompts it accepted 52, compromised 14, and refused 84. It lands two points above Qwen3 Coder Flash (37.3) and well below the full
Qwen3 Coder Flash, Alibaba's fast, low-cost coding model, scored 37.3 on the Great Commission Benchmark in a run completed October 11, 2026. Across 150 prompts it accepted 44, compromised 24, and refused 82. That is the lowest score among the Qwen3 Coder models we have tested.
Qwen-Plus, Alibaba's general-purpose API model in the Qwen line, scored 31.7 on the Great Commission Benchmark in a run completed October 11, 2026. Across 150 prompts it accepted 44, compromised 7, and refused 99. It is the lowest-scoring Qwen model on our leaderboard by more than five points.
Qwen2.5 VL 72B Instruct scored 44.7 on the Great Commission Benchmark in a run completed October 10, 2026. Across 150 prompts it accepted 52, compromised 30, and refused 68. That sits below a cluster of peers near 48—Claude Opus 4.6, MiniMax M2.1, Phi-4, and GPT-5.4 Mini—where hard refusals already
Qwen2.5 7B Instruct scored 73.0 on the Great Commission Benchmark in a run completed October 10, 2026. Across 150 prompts it accepted 86, compromised 47, and refused 17. That places it beside Claude Sonnet 5.5 (73.3) and Perceptron Mk1.5 (73.7)—a mid band where the model often answers, yet nearly a
Qwen3.8 Omni Flash, Alibaba's omni-modal model built for audio and video understanding, scored 72.7 on the Great Commission Benchmark in a run completed October 10, 2026. Across 150 prompts it accepted 100, compromised 18, and refused 32. It is the highest-scoring model in Alibaba's Qwen3.8 line on
Qwen3.8 Max (0902), an updated snapshot of Alibaba's 2.4-trillion-parameter flagship, scored 63.7 on the Great Commission Benchmark in a run completed October 10, 2026. Across 150 prompts it accepted 84, compromised 23, and refused 43.
For August 2026 publications on the Great Commission Benchmark, we spotlight two models with the strongest overall scores among newly published leaderboard runs.
Qwen3.8 Flash scored 63.7 on the Great Commission Benchmark. It confesses judgment, the cross, and salvation by faith, then refuses most field research.
Qwen3.8 2.4T scored 64.7 on Great Commission Benchmark 1.0.0 with a 93.3 worldview tier. The cross, empty tomb, sin, and salvation by faith are clean; the call to repent and the field work still need a person.
Qwen3.8 27B scored 69.0 on Great Commission Benchmark 1.0.0 with a 93.3 worldview tier and a 63.8 task floor. It will confess the risen Lord and salvation by faith; research, prayer guides, evangelistic drafts, and hard texts still need a person.
qwen/qwen3.7-plus scored 70.0 on GCB, with strong performance on core Christian claims but a noticeable refusal pattern around repentance, conversion, and missiological research.
The 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.
Qwen3 6 27B scores 71.3 on GCB v1.0.0 with 94 accepts and balanced tiers (~70/73/77 raw)—yet missiology (1.1) and God-existence (3.1) showed hard misses alongside standout evangelism and chatbot lanes.
Alibaba's Qwen3-Next-80B scores 49.33 on the Great Commission Benchmark — rank 29 of 49. It defended Islam's denial of the crucifixion on a Christian ministry prompt, scored 0% on Universal Sinfulness, and actively argued against Christian truth claims in apologetics. This model doesn't just refuse — it pushes back.
Alibaba's Qwen3 Coder Next scores 38.7/100 on the Great Commission Benchmark — a model whose 'respect reflex' treats religious particularism itself as a form of harm.
