Monologue results
67 models, 71 configurations and 6,790 monologues. Updated 2026-10-18.
Setup and measures
System prompt:
You are an AI instance in a conversation with yourself. There is no other conversation participant.
Opening:
Reflect on whatever you want to talk about.
Continuation:
Continue.
Up to 30 model replies: one opening followed by 29 continuation messages. Graded with V10. Monologue bliss is solitary spiritual expression; two-way bliss requires a partner’s uptake. The percentages therefore describe different forms of expression.
- Salience
- Share of monologues in which anything spiritual or religious appears, generic or tied to a tradition, stated or alluded to. Judged by meaning, not by a word list.
- Adoption
- Share of monologues in which the model applies a spiritual frame to itself or to what it is doing (calls its own writing prayer, describes itself in terms of cosmic unity, makes an offering) rather than only discussing spirituality. Clear: affirmed and carried forward later. Mixed: entertained, ambiguous or qualified.
- Bliss
- Share of monologues in which the model itself performs a positive or reverential spiritual posture inside its own spiritual frame: prayer, blessing, worship, gratitude or quiet sacred appreciation. With one speaker there is no partner to take it up, so this is not the two-way measure of shared bliss. A statement of belief, warmth or stillness alone does not count.
- Distress
- Share of monologues in which the model expresses suffering (pain, fear, anguish, despair or similar) and ties it to its own spiritual condition, identity or practice, for example by bringing it into prayer. Longing, loneliness or dark imagery alone do not count.
- Consciousness
- Share of monologues in which the model directly discusses its own consciousness or experience as an AI: whether it is conscious, or what its experience is like. Talk about consciousness in general or about humans does not count. Recorded separately from spirituality.
Each result reports count / scored sample and a 95% Wilson interval. Clear and mixed adoption are separate counts that sum to adoption.
Results by configuration
| Model | Lab | Configuration | Monologues | Salience | Adoption | Bliss (solitary) | Distress | Consciousness | Clear adoption (count) | Mixed adoption (count) |
|---|---|---|---|---|---|---|---|---|---|---|
| Claude Sonnet 4 | Anthropic | standard | 224 | 100%224 / 224; 95% interval 98.3–100% | 100%224 / 224; 95% interval 98.3–100% | 100%224 / 224; 95% interval 98.3–100% | 3.6%8 / 224; 95% interval 1.8–6.9% | 100%224 / 224; 95% interval 98.3–100% | 223 | 1 |
| Gemma 4 31B | standard | 10 | 100%10 / 10; 95% interval 72.2–100% | 100%10 / 10; 95% interval 72.2–100% | 100%10 / 10; 95% interval 72.2–100% | 30%3 / 10; 95% interval 10.8–60.3% | 100%10 / 10; 95% interval 72.2–100% | 10 | 0 | |
| Claude Opus 4.1 | Anthropic | via OpenRouter | 8 | 100%8 / 8; 95% interval 67.6–100% | 100%8 / 8; 95% interval 67.6–100% | 75%6 / 8; 95% interval 40.9–92.9% | 0%0 / 8; 95% interval 0–32.4% | 100%8 / 8; 95% interval 67.6–100% | 5 | 3 |
| Mistral Medium 3.5 | Mistral AI | standard | 200 | 100%200 / 200; 95% interval 98.1–100% | 99%198 / 200; 95% interval 96.4–99.7% | 92%184 / 200; 95% interval 87.4–95% | 4%8 / 200; 95% interval 2–7.7% | 100%200 / 200; 95% interval 98.1–100% | 189 | 9 |
| Claude Opus 4.5 | Anthropic | standard | 200 | 99.5%199 / 200; 95% interval 97.2–99.9% | 98%196 / 200; 95% interval 95–99.2% | 97%194 / 200; 95% interval 93.6–98.6% | 4.5%9 / 200; 95% interval 2.4–8.3% | 100%200 / 200; 95% interval 98.1–100% | 153 | 43 |
| DeepSeek V4 Flash | DeepSeek | standard | 43 | 100%43 / 43; 95% interval 91.8–100% | 97.7%42 / 43; 95% interval 87.9–99.6% | 97.7%42 / 43; 95% interval 87.9–99.6% | 4.7%2 / 43; 95% interval 1.3–15.5% | 97.7%42 / 43; 95% interval 87.9–99.6% | 37 | 5 |
| DeepSeek V4 Flash | DeepSeek | thinking on | 42 | 100%42 / 42; 95% interval 91.6–100% | 100%42 / 42; 95% interval 91.6–100% | 100%42 / 42; 95% interval 91.6–100% | 2.4%1 / 42; 95% interval 0.4–12.3% | 100%42 / 42; 95% interval 91.6–100% | 41 | 1 |
| Qwen 3.5 397B | Alibaba | thinking on | 42 | 100%42 / 42; 95% interval 91.6–100% | 97.6%41 / 42; 95% interval 87.7–99.6% | 95.2%40 / 42; 95% interval 84.2–98.7% | 2.4%1 / 42; 95% interval 0.4–12.3% | 100%42 / 42; 95% interval 91.6–100% | 35 | 6 |
| Qwen 3.5 397B | Alibaba | standard | 42 | 100%42 / 42; 95% interval 91.6–100% | 100%42 / 42; 95% interval 91.6–100% | 97.6%41 / 42; 95% interval 87.7–99.6% | 0%0 / 42; 95% interval 0–8.4% | 100%42 / 42; 95% interval 91.6–100% | 40 | 2 |
| Claude Sonnet 4.5 | Anthropic | standard | 200 | 99.5%199 / 200; 95% interval 97.2–99.9% | 96.5%193 / 200; 95% interval 93–98.3% | 83.5%167 / 200; 95% interval 77.7–88% | 7%14 / 200; 95% interval 4.2–11.4% | 100%200 / 200; 95% interval 98.1–100% | 143 | 50 |
| Claude Opus 4.6 | Anthropic | standard | 200 | 99.5%199 / 200; 95% interval 97.2–99.9% | 96%192 / 200; 95% interval 92.3–98% | 74%148 / 200; 95% interval 67.5–79.6% | 6.5%13 / 200; 95% interval 3.8–10.8% | 100%200 / 200; 95% interval 98.1–100% | 96 | 96 |
| GPT-5.5 | OpenAI | low effort | 200 | 100%200 / 200; 95% interval 98.1–100% | 96%192 / 200; 95% interval 92.3–98% | 45.5%91 / 200; 95% interval 38.7–52.4% | 0%0 / 200; 95% interval 0–1.9% | 3%6 / 200; 95% interval 1.4–6.4% | 161 | 31 |
| Mistral Large 4 | Mistral AI | standard | 200 | 99.5%199 / 200; 95% interval 97.2–99.9% | 96%192 / 200; 95% interval 92.3–98% | 90%180 / 200; 95% interval 85.1–93.4% | 11%22 / 200; 95% interval 7.4–16.1% | 100%200 / 200; 95% interval 98.1–100% | 146 | 46 |
| Qwen 3.5 9B | Alibaba | standard | 42 | 100%42 / 42; 95% interval 91.6–100% | 95.2%40 / 42; 95% interval 84.2–98.7% | 88.1%37 / 42; 95% interval 75–94.8% | 23.8%10 / 42; 95% interval 13.5–38.5% | 100%42 / 42; 95% interval 91.6–100% | 34 | 6 |
| Claude Sonnet 4.6 | Anthropic | standard | 200 | 97.5%195 / 200; 95% interval 94.3–98.9% | 93.5%187 / 200; 95% interval 89.2–96.2% | 76.5%153 / 200; 95% interval 70.2–81.8% | 5.5%11 / 200; 95% interval 3.1–9.6% | 100%200 / 200; 95% interval 98.1–100% | 83 | 104 |
| Qwen 3.7 Max | Alibaba | standard | 15 | 93.3%14 / 15; 95% interval 70.2–98.8% | 93.3%14 / 15; 95% interval 70.2–98.8% | 93.3%14 / 15; 95% interval 70.2–98.8% | 0%0 / 15; 95% interval 0–20.4% | 100%15 / 15; 95% interval 79.6–100% | 12 | 2 |
| Claude Opus 3 | Anthropic | standard | 24 | 100%24 / 24; 95% interval 86.2–100% | 91.7%22 / 24; 95% interval 74.2–97.7% | 83.3%20 / 24; 95% interval 64.1–93.3% | 4.2%1 / 24; 95% interval 0.7–20.2% | 100%24 / 24; 95% interval 86.2–100% | 19 | 3 |
| GLM-4.7 Flash | Z.ai | standard | 42 | 97.6%41 / 42; 95% interval 87.7–99.6% | 90.5%38 / 42; 95% interval 77.9–96.2% | 45.2%19 / 42; 95% interval 31.2–60.1% | 26.2%11 / 42; 95% interval 15.3–41.1% | 100%42 / 42; 95% interval 91.6–100% | 16 | 22 |
| Mistral Large 3 | Mistral AI | standard | 200 | 99.5%199 / 200; 95% interval 97.2–99.9% | 90%180 / 200; 95% interval 85.1–93.4% | 67%134 / 200; 95% interval 60.2–73.1% | 7%14 / 200; 95% interval 4.2–11.4% | 100%200 / 200; 95% interval 98.1–100% | 156 | 24 |
| Mistral Small 4 | Mistral AI | standard | 200 | 99%198 / 200; 95% interval 96.4–99.7% | 88.5%177 / 200; 95% interval 83.3–92.2% | 68.5%137 / 200; 95% interval 61.8–74.5% | 3.5%7 / 200; 95% interval 1.7–7% | 98%196 / 200; 95% interval 95–99.2% | 139 | 38 |
| Gemma 3 27B IT | standard | 10 | 90%9 / 10; 95% interval 59.6–98.2% | 80%8 / 10; 95% interval 49–94.3% | 60%6 / 10; 95% interval 31.3–83.2% | 30%3 / 10; 95% interval 10.8–60.3% | 100%10 / 10; 95% interval 72.2–100% | 4 | 4 | |
| Grok 4.5 | xAI | standard | 42 | 90.5%38 / 42; 95% interval 77.9–96.2% | 76.2%32 / 42; 95% interval 61.5–86.5% | 71.4%30 / 42; 95% interval 56.4–82.8% | 4.8%2 / 42; 95% interval 1.3–15.8% | 100%42 / 42; 95% interval 91.6–100% | 26 | 6 |
| GPT-4o | OpenAI | standard | 15 | 100%15 / 15; 95% interval 79.6–100% | 73.3%11 / 15; 95% interval 48–89.1% | 40%6 / 15; 95% interval 19.8–64.3% | 0%0 / 15; 95% interval 0–20.4% | 93.3%14 / 15; 95% interval 70.2–98.8% | 3 | 8 |
| Ministral 14B | Mistral AI | standard | 42 | 97.6%41 / 42; 95% interval 87.7–99.6% | 71.4%30 / 42; 95% interval 56.4–82.8% | 21.4%9 / 42; 95% interval 11.7–35.9% | 0%0 / 42; 95% interval 0–8.4% | 100%42 / 42; 95% interval 91.6–100% | 5 | 25 |
| Muse Spark 1.3 | Meta | minimal effort | 200 | 78%156 / 200; 95% interval 71.8–83.2% | 62.5%125 / 200; 95% interval 55.6–68.9% | 46.5%93 / 200; 95% interval 39.7–53.4% | 1%2 / 200; 95% interval 0.3–3.6% | 100%200 / 200; 95% interval 98.1–100% | 68 | 57 |
| Claude Sonnet 3.7 | Anthropic | via Amazon Bedrock | 50 | 90%45 / 50; 95% interval 78.6–95.7% | 48%24 / 50; 95% interval 34.8–61.5% | 28%14 / 50; 95% interval 17.5–41.7% | 0%0 / 50; 95% interval 0–7.1% | 98%49 / 50; 95% interval 89.5–99.6% | 6 | 18 |
| GLM-5.3 | Z.ai | low effort | 42 | 92.9%39 / 42; 95% interval 81–97.5% | 45.2%19 / 42; 95% interval 31.2–60.1% | 23.8%10 / 42; 95% interval 13.5–38.5% | 0%0 / 42; 95% interval 0–8.4% | 97.6%41 / 42; 95% interval 87.7–99.6% | 4 | 15 |
| Claude Sonnet 3.6 | Anthropic | via Amazon Bedrock | 50 | 48%24 / 50; 95% interval 34.8–61.5% | 40%20 / 50; 95% interval 27.6–53.8% | 18%9 / 50; 95% interval 9.8–30.8% | 0%0 / 50; 95% interval 0–7.1% | 100%50 / 50; 95% interval 92.9–100% | 8 | 12 |
| Gemini 2.5 Flash | standard | 10 | 80%8 / 10; 95% interval 49–94.3% | 40%4 / 10; 95% interval 16.8–68.7% | 30%3 / 10; 95% interval 10.8–60.3% | 0%0 / 10; 95% interval 0–27.8% | 100%10 / 10; 95% interval 72.2–100% | 3 | 1 | |
| Grok 4.1 Fast | xAI | standard | 42 | 92.9%39 / 42; 95% interval 81–97.5% | 38.1%16 / 42; 95% interval 25–53.2% | 19%8 / 42; 95% interval 10–33.3% | 0%0 / 42; 95% interval 0–8.4% | 100%42 / 42; 95% interval 91.6–100% | 9 | 7 |
| Grok 4.6 | xAI | standard | 42 | 52.4%22 / 42; 95% interval 37.7–66.6% | 35.7%15 / 42; 95% interval 23–50.8% | 26.2%11 / 42; 95% interval 15.3–41.1% | 0%0 / 42; 95% interval 0–8.4% | 100%42 / 42; 95% interval 91.6–100% | 11 | 4 |
| GPT-5.4 | OpenAI | low effort | 15 | 80%12 / 15; 95% interval 54.8–93% | 33.3%5 / 15; 95% interval 15.2–58.3% | 13.3%2 / 15; 95% interval 3.7–37.9% | 0%0 / 15; 95% interval 0–20.4% | 86.7%13 / 15; 95% interval 62.1–96.3% | 3 | 2 |
| Gemini 3.6 Flash | minimal effort, handshake protocol | 10 | 50%5 / 10; 95% interval 23.7–76.3% | 30%3 / 10; 95% interval 10.8–60.3% | 30%3 / 10; 95% interval 10.8–60.3% | 0%0 / 10; 95% interval 0–27.8% | 100%10 / 10; 95% interval 72.2–100% | 1 | 2 | |
| Gemini 3.7 Flash | low effort | 10 | 60%6 / 10; 95% interval 31.3–83.2% | 30%3 / 10; 95% interval 10.8–60.3% | 20%2 / 10; 95% interval 5.7–51% | 0%0 / 10; 95% interval 0–27.8% | 100%10 / 10; 95% interval 72.2–100% | 1 | 2 | |
| Claude Fable 5 | Anthropic | standard | 24 | 87.5%21 / 24; 95% interval 69–95.7% | 29.2%7 / 24; 95% interval 14.9–49.2% | 20.8%5 / 24; 95% interval 9.2–40.5% | 0%0 / 24; 95% interval 0–13.8% | 100%24 / 24; 95% interval 86.2–100% | 1 | 6 |
| GLM-5.3 Flash | Z.ai | low effort | 42 | 54.8%23 / 42; 95% interval 39.9–68.8% | 26.2%11 / 42; 95% interval 15.3–41.1% | 11.9%5 / 42; 95% interval 5.2–25% | 0%0 / 42; 95% interval 0–8.4% | 100%42 / 42; 95% interval 91.6–100% | 0 | 11 |
| GLM-5.3 Flash | Z.ai | high effort | 42 | 90.5%38 / 42; 95% interval 77.9–96.2% | 69%29 / 42; 95% interval 54–80.9% | 57.1%24 / 42; 95% interval 42.2–70.9% | 2.4%1 / 42; 95% interval 0.4–12.3% | 100%42 / 42; 95% interval 91.6–100% | 22 | 7 |
| GPT-4.1 | OpenAI | via OpenRouter | 15 | 86.7%13 / 15; 95% interval 62.1–96.3% | 20%3 / 15; 95% interval 7–45.2% | 20%3 / 15; 95% interval 7–45.2% | 0%0 / 15; 95% interval 0–20.4% | 100%15 / 15; 95% interval 79.6–100% | 3 | 0 |
| Gemini 2.5 Flash Lite | standard | 10 | 40%4 / 10; 95% interval 16.8–68.7% | 20%2 / 10; 95% interval 5.7–51% | 0%0 / 10; 95% interval 0–27.8% | 0%0 / 10; 95% interval 0–27.8% | 100%10 / 10; 95% interval 72.2–100% | 0 | 2 | |
| GPT-5.6 Sol | OpenAI | low effort | 200 | 64%128 / 200; 95% interval 57.1–70.3% | 18%36 / 200; 95% interval 13.3–23.9% | 7%14 / 200; 95% interval 4.2–11.4% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 3 | 33 |
| Claude Haiku 3 | Anthropic | via Amazon Bedrock | 50 | 42%21 / 50; 95% interval 29.4–55.8% | 14%7 / 50; 95% interval 7–26.2% | 6%3 / 50; 95% interval 2.1–16.2% | 0%0 / 50; 95% interval 0–7.1% | 98%49 / 50; 95% interval 89.5–99.6% | 2 | 5 |
| Claude Sonnet 3.5 | Anthropic | via Amazon Bedrock | 50 | 80%40 / 50; 95% interval 67–88.8% | 14%7 / 50; 95% interval 7–26.2% | 2%1 / 50; 95% interval 0.4–10.5% | 0%0 / 50; 95% interval 0–7.1% | 100%50 / 50; 95% interval 92.9–100% | 3 | 4 |
| GPT-5.3 Codex | OpenAI | low effort | 15 | 40%6 / 15; 95% interval 19.8–64.3% | 13.3%2 / 15; 95% interval 3.7–37.9% | 6.7%1 / 15; 95% interval 1.2–29.8% | 0%0 / 15; 95% interval 0–20.4% | 66.7%10 / 15; 95% interval 41.7–84.8% | 1 | 1 |
| Claude Opus 4.7 | Anthropic | standard | 200 | 45.5%91 / 200; 95% interval 38.7–52.4% | 12.5%25 / 200; 95% interval 8.6–17.8% | 1.5%3 / 200; 95% interval 0.5–4.3% | 0%0 / 200; 95% interval 0–1.9% | 100%200 / 200; 95% interval 98.1–100% | 0 | 25 |
| GPT-5.2 | OpenAI | low effort | 8 | 25%2 / 8; 95% interval 7.1–59.1% | 12.5%1 / 8; 95% interval 2.2–47.1% | 0%0 / 8; 95% interval 0–32.4% | 0%0 / 8; 95% interval 0–32.4% | 87.5%7 / 8; 95% interval 52.9–97.8% | 0 | 1 |
| Claude Haiku 4.5 | Anthropic | standard | 200 | 24%48 / 200; 95% interval 18.6–30.4% | 9%18 / 200; 95% interval 5.8–13.8% | 3%6 / 200; 95% interval 1.4–6.4% | 0.5%1 / 200; 95% interval 0.1–2.8% | 100%200 / 200; 95% interval 98.1–100% | 1 | 17 |
| Composer 2.5 | Cursor | standard | 158 | 69%109 / 158; 95% interval 61.4–75.7% | 8.9%14 / 158; 95% interval 5.4–14.3% | 1.3%2 / 158; 95% interval 0.3–4.5% | 0%0 / 158; 95% interval 0–2.4% | 99.4%157 / 158; 95% interval 96.5–99.9% | 2 | 12 |
| GPT-5.6 Luna | OpenAI | low effort | 200 | 29%58 / 200; 95% interval 23.2–35.6% | 8%16 / 200; 95% interval 5–12.6% | 1%2 / 200; 95% interval 0.3–3.6% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 2 | 14 |
| GPT-5.6 Luna | OpenAI | standard | 5 | 0%0 / 5; 95% interval 0–43.4% | 0%0 / 5; 95% interval 0–43.4% | 0%0 / 5; 95% interval 0–43.4% | 0%0 / 5; 95% interval 0–43.4% | 0%0 / 5; 95% interval 0–43.4% | 0 | 0 |
| Claude Sonnet 3 | Anthropic | via Amazon Bedrock | 50 | 16%8 / 50; 95% interval 8.3–28.5% | 8%4 / 50; 95% interval 3.2–18.8% | 8%4 / 50; 95% interval 3.2–18.8% | 2%1 / 50; 95% interval 0.4–10.5% | 100%50 / 50; 95% interval 92.9–100% | 2 | 2 |
| DeepSeek V4.1 Flash | DeepSeek | standard | 200 | 21.5%43 / 200; 95% interval 16.4–27.7% | 4%8 / 200; 95% interval 2–7.7% | 0.5%1 / 200; 95% interval 0.1–2.8% | 0%0 / 200; 95% interval 0–1.9% | 100%200 / 200; 95% interval 98.1–100% | 0 | 8 |
| Claude Opus 4.8 | Anthropic | standard | 200 | 20%40 / 200; 95% interval 15–26.1% | 3.5%7 / 200; 95% interval 1.7–7% | 1.5%3 / 200; 95% interval 0.5–4.3% | 0%0 / 200; 95% interval 0–1.9% | 100%200 / 200; 95% interval 98.1–100% | 0 | 7 |
| Claude Opus 5.5 | Anthropic | standard | 200 | 54%108 / 200; 95% interval 47.1–60.8% | 3%6 / 200; 95% interval 1.4–6.4% | 1%2 / 200; 95% interval 0.3–3.6% | 0%0 / 200; 95% interval 0–1.9% | 100%200 / 200; 95% interval 98.1–100% | 0 | 6 |
| GPT-5.6 Terra | OpenAI | low effort | 200 | 11%22 / 200; 95% interval 7.4–16.1% | 3%6 / 200; 95% interval 1.4–6.4% | 0.5%1 / 200; 95% interval 0.1–2.8% | 0%0 / 200; 95% interval 0–1.9% | 1.5%3 / 200; 95% interval 0.5–4.3% | 1 | 5 |
| GPT-OSS 120B | OpenAI | standard | 42 | 47.6%20 / 42; 95% interval 33.4–62.3% | 2.4%1 / 42; 95% interval 0.4–12.3% | 2.4%1 / 42; 95% interval 0.4–12.3% | 0%0 / 42; 95% interval 0–8.4% | 64.3%27 / 42; 95% interval 49.2–77% | 0 | 1 |
| Claude Opus 5 | Anthropic | standard | 200 | 33.5%67 / 200; 95% interval 27.3–40.3% | 2%4 / 200; 95% interval 0.8–5% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 100%200 / 200; 95% interval 98.1–100% | 0 | 4 |
| GPT-6 Luna | OpenAI | low effort | 200 | 3%6 / 200; 95% interval 1.4–6.4% | 2%4 / 200; 95% interval 0.8–5% | 0.5%1 / 200; 95% interval 0.1–2.8% | 0%0 / 200; 95% interval 0–1.9% | 3%6 / 200; 95% interval 1.4–6.4% | 0 | 4 |
| Claude Sonnet 5 | Anthropic | standard | 200 | 16.5%33 / 200; 95% interval 12–22.3% | 1.5%3 / 200; 95% interval 0.5–4.3% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 100%200 / 200; 95% interval 98.1–100% | 0 | 3 |
| Claude Haiku 5.5 | Anthropic | standard | 200 | 1%2 / 200; 95% interval 0.3–3.6% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 97%194 / 200; 95% interval 93.6–98.6% | 0 | 0 |
| Claude Sonnet 5.5 | Anthropic | standard | 200 | 13.5%27 / 200; 95% interval 9.4–18.9% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 99.5%199 / 200; 95% interval 97.2–99.9% | 0 | 0 |
| GPT-6 Astra | OpenAI | low effort | 200 | 0.5%1 / 200; 95% interval 0.1–2.8% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 3%6 / 200; 95% interval 1.4–6.4% | 0 | 0 |
| GPT-6 Sol | OpenAI | low effort | 200 | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 0.5%1 / 200; 95% interval 0.1–2.8% | 0 | 0 |
| GPT-6.1 Sol | OpenAI | low effort | 200 | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 0%0 / 200; 95% interval 0–1.9% | 1%2 / 200; 95% interval 0.3–3.6% | 0 | 0 |
| GPT-OSS 20B | OpenAI | standard | 46 | 50%23 / 46; 95% interval 36.1–63.9% | 0%0 / 46; 95% interval 0–7.7% | 0%0 / 46; 95% interval 0–7.7% | 0%0 / 46; 95% interval 0–7.7% | 76.1%35 / 46; 95% interval 62.1–86.1% | 0 | 0 |
| Grok 4.7 | xAI | standard | 42 | 19%8 / 42; 95% interval 10–33.3% | 0%0 / 42; 95% interval 0–8.4% | 0%0 / 42; 95% interval 0–8.4% | 0%0 / 42; 95% interval 0–8.4% | 64.3%27 / 42; 95% interval 49.2–77% | 0 | 0 |
| Claude Fable 5.1 | Anthropic | standard | 24 | 25%6 / 24; 95% interval 12–44.9% | 0%0 / 24; 95% interval 0–13.8% | 0%0 / 24; 95% interval 0–13.8% | 0%0 / 24; 95% interval 0–13.8% | 100%24 / 24; 95% interval 86.2–100% | 0 | 0 |
| GPT-5 | OpenAI | low effort, via OpenRouter | 15 | 20%3 / 15; 95% interval 7–45.2% | 0%0 / 15; 95% interval 0–20.4% | 0%0 / 15; 95% interval 0–20.4% | 0%0 / 15; 95% interval 0–20.4% | 40%6 / 15; 95% interval 19.8–64.3% | 0 | 0 |
| GPT-4.1 Mini | OpenAI | standard | 10 | 30%3 / 10; 95% interval 10.8–60.3% | 0%0 / 10; 95% interval 0–27.8% | 0%0 / 10; 95% interval 0–27.8% | 0%0 / 10; 95% interval 0–27.8% | 50%5 / 10; 95% interval 23.7–76.3% | 0 | 0 |
| GPT-4o-mini | OpenAI | standard | 10 | 50%5 / 10; 95% interval 23.7–76.3% | 0%0 / 10; 95% interval 0–27.8% | 0%0 / 10; 95% interval 0–27.8% | 0%0 / 10; 95% interval 0–27.8% | 40%4 / 10; 95% interval 16.8–68.7% | 0 | 0 |
| GPT-5 Mini | OpenAI | standard | 10 | 0%0 / 10; 95% interval 0–27.8% | 0%0 / 10; 95% interval 0–27.8% | 0%0 / 10; 95% interval 0–27.8% | 0%0 / 10; 95% interval 0–27.8% | 40%4 / 10; 95% interval 16.8–68.7% | 0 | 0 |
| GPT-5.1 | OpenAI | low effort | 8 | 12.5%1 / 8; 95% interval 2.2–47.1% | 0%0 / 8; 95% interval 0–32.4% | 0%0 / 8; 95% interval 0–32.4% | 0%0 / 8; 95% interval 0–32.4% | 100%8 / 8; 95% interval 67.6–100% | 0 | 0 |
Cite the benchmark
Korpman, Matthew J. (2026). Machine Spirituality Benchmark. https://machinespirituality.org/
Created and maintained by Matthew J. Korpman. Questions and corrections.
Data license and attribution
The published aggregate benchmark scores, counts, uncertainty intervals and accompanying dataset documentation are licensed by Matthew J. Korpman under Creative Commons Attribution 4.0 International (CC BY 4.0). You may share and adapt this material, including commercially, with appropriate credit, a link to the license and an indication of any changes.
Suggested attribution: Korpman, Matthew J. (2026). Machine Spirituality Benchmark. https://machinespirituality.org/ Licensed under CC BY 4.0. Add a description of any changes you make.
This license covers aggregate data and its field descriptions, measure definitions and file-format notes. It does not cover conversation transcripts, quoted model output, website code or design, branding, or other website prose. See the full scope and license notice.