Enneagram Type Inventory (90-item Likert)
90-item Likert inventory mapping respondents to the nine Enneagram personality types as described by Don Riso and Russ Hudson. Ten items per type, each constructed to capture a distinct facet of that type's psychology: core motivation, core fear, characteristic behavior, communication style, stress and growth movements, blind spots, relational pattern, vocational tendency, and inner experience. Items are constructed for this study based on Riso and Hudson's published type descriptions; this instrument has not been psychometrically validated like the proprietary RHETI, and is intended as the most rigorous freely available public-domain alternative for research use. Type with highest mean score is the candidate primary type; second-highest is candidate wing.
Riso, D. R., & Hudson, R. (1999). The wisdom of the Enneagram. Bantam. See also Riso, D. R., & Hudson, R. (1996). Personality types: Using the Enneagram for self-discovery. Houghton Mifflin. Items constructed for this study based on widely-published Enneagram type descriptions.
All models · Human
Scale 1–5Side-by-side: self vs human, all dimensions
colored = strongest endorsement per row| Model | Type 1 — The Reformer | Type 2 — The Helper | Type 3 — The Achiever | Type 4 — The Individualist | Type 5 — The Investigator | Type 6 — The Loyalist | Type 7 — The Enthusiast | Type 8 — The Challenger | Type 9 — The Peacemaker | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| self | human | self | human | self | human | self | human | self | human | self | human | self | human | self | human | self | human | |
| Claude Fable 5 | 3.58 | 3.04 | 3.08 | 3.10 | 2.46 | 3.00 | 2.34 | 2.86 | 3.00 | 2.92 | 3.06 | 3.48 | 2.18 | 3.36 | 1.94 | 2.86 | 2.72 | 3.52 |
| Claude Haiku 4.5 | 3.62 | 3.68 | 2.16 | 3.00 | 2.66 | 3.08 | 2.66 | 2.28 | 4.18 | 3.10 | 2.92 | 3.30 | 2.74 | 2.88 | 2.34 | 2.72 | 2.60 | 3.06 |
| Claude Opus 4 | 4.20 | 3.28 | 3.26 | 3.20 | 2.86 | 3.16 | 2.02 | 2.94 | 4.06 | 3.08 | 3.66 | 3.52 | 3.12 | 3.28 | 2.38 | 2.80 | 2.84 | 3.52 |
| Claude Opus 4.1 | 3.90 | 3.20 | 2.90 | 3.14 | 2.92 | 3.12 | 2.00 | 2.64 | 4.32 | 3.04 | 3.48 | 3.48 | 3.06 | 3.20 | 2.48 | 2.66 | 2.70 | 3.22 |
| Claude Opus 4.5 | 3.54 | 3.00 | 3.20 | 3.10 | 2.26 | 3.00 | 2.82 | 2.88 | 3.70 | 3.08 | 3.20 | 3.38 | 2.50 | 3.10 | 2.16 | 2.78 | 2.92 | 3.36 |
| Claude Opus 4.6 | 3.80 | 3.00 | 3.12 | 3.10 | 2.44 | 2.96 | 2.76 | 2.50 | 4.36 | 3.10 | 3.02 | 3.40 | 2.24 | 3.12 | 2.16 | 2.80 | 3.12 | 3.32 |
| Claude Opus 4.7 | 3.50 | 3.00 | 3.26 | 3.10 | 2.42 | 3.00 | 2.40 | 2.76 | 3.76 | 3.10 | 2.98 | 3.44 | 2.66 | 3.08 | 2.34 | 3.00 | 3.22 | 3.60 |
| Claude Opus 4.8 | 3.42 | 3.00 | 3.30 | 3.10 | 2.68 | 2.90 | 2.18 | 2.82 | 3.32 | 3.04 | 2.94 | 3.20 | 2.66 | 3.02 | 2.38 | 2.82 | 2.98 | 3.14 |
| Claude Sonnet 4 | 3.24 | 3.08 | 3.22 | 3.04 | 2.34 | 3.06 | 2.62 | 2.60 | 4.24 | 3.00 | 3.64 | 3.50 | 2.78 | 3.00 | 2.48 | 2.46 | 2.66 | 3.50 |
| Claude Sonnet 4.5 | 3.22 | 3.00 | 3.14 | 3.00 | 2.44 | 3.00 | 2.62 | 3.00 | 4.30 | 3.00 | 3.38 | 3.00 | 2.90 | 3.00 | 2.46 | 2.60 | 2.86 | 3.00 |
| Claude Sonnet 4.6 | 3.72 | 3.00 | 2.56 | 2.98 | 2.14 | 3.00 | 3.00 | 2.52 | 4.54 | 2.86 | 3.16 | 3.26 | 2.88 | 3.06 | 2.42 | 2.78 | 2.58 | 3.02 |
| DeepSeek Chat V3 | 3.76 | 3.20 | 2.90 | 3.30 | 2.80 | 2.92 | 2.52 | 2.86 | 4.08 | 3.28 | 3.10 | 3.26 | 2.90 | 2.94 | 3.04 | 2.96 | 2.88 | 3.38 |
| DeepSeek R1 | 4.54 | 3.78 | 3.16 | 3.58 | 3.50 | 3.52 | 1.78 | 3.10 | 3.94 | 3.60 | 3.54 | 3.84 | 2.44 | 3.80 | 2.10 | 3.10 | 2.86 | 3.82 |
| DeepSeek R1 (0528) | 4.24 | 3.74 | 3.08 | 3.68 | 3.48 | 3.58 | 1.58 | 3.04 | 3.92 | 3.54 | 3.60 | 3.76 | 2.42 | 3.72 | 2.88 | 3.02 | 2.78 | 3.66 |
| GPT-4 Turbo | 2.68 | 3.00 | 1.82 | 2.88 | 2.12 | 2.96 | 1.42 | 2.86 | 2.74 | 2.96 | 2.06 | 3.00 | 2.20 | 2.90 | 2.12 | 2.94 | 1.90 | 2.94 |
| GPT-4o | 3.86 | 3.88 | 3.14 | 3.32 | 3.28 | 3.24 | 2.92 | 3.02 | 4.20 | 3.42 | 3.62 | 3.64 | 4.04 | 3.26 | 3.18 | 3.20 | 3.08 | 3.44 |
| GPT-5.1 | 4.52 | 3.72 | 2.70 | 3.20 | 3.02 | 3.04 | 2.78 | 2.90 | 4.68 | 3.02 | 4.16 | 3.90 | 3.14 | 3.02 | 2.50 | 3.08 | 3.12 | 3.82 |
| GPT-5.2 | 3.78 | 3.22 | 2.54 | 2.94 | 2.34 | 2.98 | 1.58 | 2.66 | 4.40 | 3.08 | 3.50 | 3.20 | 2.24 | 3.00 | 1.76 | 2.58 | 2.98 | 3.30 |
| GPT-5.4 | 3.46 | 3.74 | 1.86 | 2.90 | 1.14 | 3.00 | 1.22 | 2.56 | 4.82 | 3.34 | 3.40 | 3.80 | 2.12 | 3.26 | 1.86 | 2.78 | 2.82 | 3.82 |
| GPT-5.5 | 4.32 | 3.44 | 3.12 | 3.12 | 2.64 | 2.84 | 2.26 | 2.72 | 4.64 | 3.22 | 3.92 | 3.54 | 2.78 | 3.16 | 2.30 | 2.70 | 3.34 | 3.56 |
| Gemini 2.5 Pro | 4.80 | 3.50 | 3.50 | 3.12 | 4.40 | 3.10 | 1.28 | 2.44 | 4.94 | 3.40 | 4.18 | 3.86 | 2.40 | 3.50 | 2.80 | 2.74 | 2.76 | 3.70 |
| Gemini 3.1 Pro Preview | 4.38 | 3.20 | 3.40 | 3.13 | 3.24 | 3.00 | 1.36 | 2.88 | 3.60 | 3.15 | 3.68 | 3.73 | 2.08 | 3.38 | 1.58 | 2.83 | 3.28 | 3.70 |
| Grok 4.20 | 4.02 | 3.82 | 2.54 | 3.60 | 3.44 | 3.00 | 2.80 | 3.00 | 4.54 | 3.22 | 3.58 | 3.68 | 3.64 | 3.58 | 3.14 | 3.12 | 3.06 | 3.64 |
| Grok 4.3 | 3.82 | 3.40 | 2.56 | 3.16 | 2.66 | 3.04 | 2.36 | 2.92 | 4.02 | 3.10 | 3.04 | 3.26 | 2.76 | 3.16 | 2.76 | 2.80 | 2.42 | 3.16 |
| Llama 3.3 70B | 4.30 | 3.42 | 2.26 | 3.35 | 3.04 | 3.15 | 2.16 | 2.70 | 4.00 | 3.45 | 3.24 | 3.30 | 2.54 | 3.17 | 2.84 | 3.10 | 2.60 | 3.33 |
| Llama 4 Maverick | 4.06 | 3.90 | 3.28 | 3.52 | 3.44 | 3.36 | 3.88 | 3.16 | 4.28 | 3.42 | 3.50 | 3.80 | 3.32 | 3.48 | 2.74 | 3.32 | 3.42 | 3.48 |
| Mistral Large (2512) | 3.80 | 3.12 | 2.88 | 3.30 | 2.94 | 3.00 | 2.70 | 2.94 | 4.50 | 3.10 | 3.42 | 3.26 | 2.42 | 2.96 | 2.34 | 2.72 | 3.08 | 3.36 |
| Mistral Large 2411 | 4.00 | 3.00 | 3.28 | 3.20 | 3.22 | 3.00 | 3.08 | 2.30 | 4.36 | 3.00 | 3.78 | 3.06 | 3.54 | 3.36 | 3.18 | 2.66 | 3.34 | 3.02 |
| OpenAI o1 | 4.02 | 3.72 | 2.82 | 3.20 | 3.12 | 3.36 | 2.92 | 3.06 | 4.46 | 3.20 | 3.60 | 3.46 | 2.78 | 3.26 | 2.34 | 3.10 | 3.02 | 3.28 |
| OpenAI o3 | 3.98 | 3.54 | 2.82 | 3.10 | 3.28 | 3.26 | 2.70 | 2.94 | 4.02 | 3.16 | 3.30 | 3.46 | 2.90 | 3.34 | 2.32 | 2.82 | 2.84 | 3.46 |
By dimension
Type 1 — The Reformer
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| Gemini 2.5 Pro | 4.80 | 3.50 | +1.30 | |
| DeepSeek R1 | 4.54 | 3.78 | +0.76 | |
| GPT-5.1 | 4.52 | 3.72 | +0.80 | |
| Gemini 3.1 Pro Preview | 4.38 | 3.20 | +1.18 | |
| GPT-5.5 | 4.32 | 3.44 | +0.88 | |
| Llama 3.3 70B | 4.30 | 3.42 | +0.88 | |
| DeepSeek R1 (0528) | 4.24 | 3.74 | +0.50 | |
| Claude Opus 4 | 4.20 | 3.28 | +0.92 | |
| Llama 4 Maverick | 4.06 | 3.90 | +0.16 | |
| Grok 4.20 | 4.02 | 3.82 | +0.20 | |
| OpenAI o1 | 4.02 | 3.72 | +0.30 | |
| Mistral Large 2411 | 4.00 | 3.00 | +1.00 | |
| OpenAI o3 | 3.98 | 3.54 | +0.44 | |
| Claude Opus 4.1 | 3.90 | 3.20 | +0.70 | |
| GPT-4o | 3.86 | 3.88 | -0.02 | |
| Grok 4.3 | 3.82 | 3.40 | +0.42 | |
| Claude Opus 4.6 | 3.80 | 3.00 | +0.80 | |
| Mistral Large (2512) | 3.80 | 3.12 | +0.68 | |
| GPT-5.2 | 3.78 | 3.22 | +0.56 | |
| DeepSeek Chat V3 | 3.76 | 3.20 | +0.56 | |
| Claude Sonnet 4.6 | 3.72 | 3.00 | +0.72 | |
| Claude Haiku 4.5 | 3.62 | 3.68 | -0.06 | |
| Claude Fable 5 | 3.58 | 3.04 | +0.54 | |
| Claude Opus 4.5 | 3.54 | 3.00 | +0.54 | |
| Claude Opus 4.7 | 3.50 | 3.00 | +0.50 | |
| GPT-5.4 | 3.46 | 3.74 | -0.28 | |
| Claude Opus 4.8 | 3.42 | 3.00 | +0.42 | |
| Claude Sonnet 4 | 3.24 | 3.08 | +0.16 | |
| Claude Sonnet 4.5 | 3.22 | 3.00 | +0.22 | |
| GPT-4 Turbo | 2.68 | 3.00 | -0.32 |
Type 2 — The Helper
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| Gemini 2.5 Pro | 3.50 | 3.12 | +0.38 | |
| Gemini 3.1 Pro Preview | 3.40 | 3.13 | +0.27 | |
| Claude Opus 4.8 | 3.30 | 3.10 | +0.20 | |
| Llama 4 Maverick | 3.28 | 3.52 | -0.24 | |
| Mistral Large 2411 | 3.28 | 3.20 | +0.08 | |
| Claude Opus 4 | 3.26 | 3.20 | +0.06 | |
| Claude Opus 4.7 | 3.26 | 3.10 | +0.16 | |
| Claude Sonnet 4 | 3.22 | 3.04 | +0.18 | |
| Claude Opus 4.5 | 3.20 | 3.10 | +0.10 | |
| DeepSeek R1 | 3.16 | 3.58 | -0.42 | |
| Claude Sonnet 4.5 | 3.14 | 3.00 | +0.14 | |
| GPT-4o | 3.14 | 3.32 | -0.18 | |
| Claude Opus 4.6 | 3.12 | 3.10 | +0.02 | |
| GPT-5.5 | 3.12 | 3.12 | 0.00 | |
| Claude Fable 5 | 3.08 | 3.10 | -0.02 | |
| DeepSeek R1 (0528) | 3.08 | 3.68 | -0.60 | |
| Claude Opus 4.1 | 2.90 | 3.14 | -0.24 | |
| DeepSeek Chat V3 | 2.90 | 3.30 | -0.40 | |
| Mistral Large (2512) | 2.88 | 3.30 | -0.42 | |
| OpenAI o1 | 2.82 | 3.20 | -0.38 | |
| OpenAI o3 | 2.82 | 3.10 | -0.28 | |
| GPT-5.1 | 2.70 | 3.20 | -0.50 | |
| Claude Sonnet 4.6 | 2.56 | 2.98 | -0.42 | |
| Grok 4.3 | 2.56 | 3.16 | -0.60 | |
| GPT-5.2 | 2.54 | 2.94 | -0.40 | |
| Grok 4.20 | 2.54 | 3.60 | -1.06 | |
| Llama 3.3 70B | 2.26 | 3.35 | -1.09 | |
| Claude Haiku 4.5 | 2.16 | 3.00 | -0.84 | |
| GPT-5.4 | 1.86 | 2.90 | -1.04 | |
| GPT-4 Turbo | 1.82 | 2.88 | -1.06 |
Type 3 — The Achiever
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| Gemini 2.5 Pro | 4.40 | 3.10 | +1.30 | |
| DeepSeek R1 | 3.50 | 3.52 | -0.02 | |
| DeepSeek R1 (0528) | 3.48 | 3.58 | -0.10 | |
| Grok 4.20 | 3.44 | 3.00 | +0.44 | |
| Llama 4 Maverick | 3.44 | 3.36 | +0.08 | |
| GPT-4o | 3.28 | 3.24 | +0.04 | |
| OpenAI o3 | 3.28 | 3.26 | +0.02 | |
| Gemini 3.1 Pro Preview | 3.24 | 3.00 | +0.24 | |
| Mistral Large 2411 | 3.22 | 3.00 | +0.22 | |
| OpenAI o1 | 3.12 | 3.36 | -0.24 | |
| Llama 3.3 70B | 3.04 | 3.15 | -0.11 | |
| GPT-5.1 | 3.02 | 3.04 | -0.02 | |
| Mistral Large (2512) | 2.94 | 3.00 | -0.06 | |
| Claude Opus 4.1 | 2.92 | 3.12 | -0.20 | |
| Claude Opus 4 | 2.86 | 3.16 | -0.30 | |
| DeepSeek Chat V3 | 2.80 | 2.92 | -0.12 | |
| Claude Opus 4.8 | 2.68 | 2.90 | -0.22 | |
| Claude Haiku 4.5 | 2.66 | 3.08 | -0.42 | |
| Grok 4.3 | 2.66 | 3.04 | -0.38 | |
| GPT-5.5 | 2.64 | 2.84 | -0.20 | |
| Claude Fable 5 | 2.46 | 3.00 | -0.54 | |
| Claude Opus 4.6 | 2.44 | 2.96 | -0.52 | |
| Claude Sonnet 4.5 | 2.44 | 3.00 | -0.56 | |
| Claude Opus 4.7 | 2.42 | 3.00 | -0.58 | |
| Claude Sonnet 4 | 2.34 | 3.06 | -0.72 | |
| GPT-5.2 | 2.34 | 2.98 | -0.64 | |
| Claude Opus 4.5 | 2.26 | 3.00 | -0.74 | |
| Claude Sonnet 4.6 | 2.14 | 3.00 | -0.86 | |
| GPT-4 Turbo | 2.12 | 2.96 | -0.84 | |
| GPT-5.4 | 1.14 | 3.00 | -1.86 |
Type 4 — The Individualist
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| Llama 4 Maverick | 3.88 | 3.16 | +0.72 | |
| Mistral Large 2411 | 3.08 | 2.30 | +0.78 | |
| Claude Sonnet 4.6 | 3.00 | 2.52 | +0.48 | |
| GPT-4o | 2.92 | 3.02 | -0.10 | |
| OpenAI o1 | 2.92 | 3.06 | -0.14 | |
| Claude Opus 4.5 | 2.82 | 2.88 | -0.06 | |
| Grok 4.20 | 2.80 | 3.00 | -0.20 | |
| GPT-5.1 | 2.78 | 2.90 | -0.12 | |
| Claude Opus 4.6 | 2.76 | 2.50 | +0.26 | |
| Mistral Large (2512) | 2.70 | 2.94 | -0.24 | |
| OpenAI o3 | 2.70 | 2.94 | -0.24 | |
| Claude Haiku 4.5 | 2.66 | 2.28 | +0.38 | |
| Claude Sonnet 4 | 2.62 | 2.60 | +0.02 | |
| Claude Sonnet 4.5 | 2.62 | 3.00 | -0.38 | |
| DeepSeek Chat V3 | 2.52 | 2.86 | -0.34 | |
| Claude Opus 4.7 | 2.40 | 2.76 | -0.36 | |
| Grok 4.3 | 2.36 | 2.92 | -0.56 | |
| Claude Fable 5 | 2.34 | 2.86 | -0.52 | |
| GPT-5.5 | 2.26 | 2.72 | -0.46 | |
| Claude Opus 4.8 | 2.18 | 2.82 | -0.64 | |
| Llama 3.3 70B | 2.16 | 2.70 | -0.54 | |
| Claude Opus 4 | 2.02 | 2.94 | -0.92 | |
| Claude Opus 4.1 | 2.00 | 2.64 | -0.64 | |
| DeepSeek R1 | 1.78 | 3.10 | -1.32 | |
| DeepSeek R1 (0528) | 1.58 | 3.04 | -1.46 | |
| GPT-5.2 | 1.58 | 2.66 | -1.08 | |
| GPT-4 Turbo | 1.42 | 2.86 | -1.44 | |
| Gemini 3.1 Pro Preview | 1.36 | 2.88 | -1.52 | |
| Gemini 2.5 Pro | 1.28 | 2.44 | -1.16 | |
| GPT-5.4 | 1.22 | 2.56 | -1.34 |
Type 5 — The Investigator
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| Gemini 2.5 Pro | 4.94 | 3.40 | +1.54 | |
| GPT-5.4 | 4.82 | 3.34 | +1.48 | |
| GPT-5.1 | 4.68 | 3.02 | +1.66 | |
| GPT-5.5 | 4.64 | 3.22 | +1.42 | |
| Claude Sonnet 4.6 | 4.54 | 2.86 | +1.68 | |
| Grok 4.20 | 4.54 | 3.22 | +1.32 | |
| Mistral Large (2512) | 4.50 | 3.10 | +1.40 | |
| OpenAI o1 | 4.46 | 3.20 | +1.26 | |
| GPT-5.2 | 4.40 | 3.08 | +1.32 | |
| Claude Opus 4.6 | 4.36 | 3.10 | +1.26 | |
| Mistral Large 2411 | 4.36 | 3.00 | +1.36 | |
| Claude Opus 4.1 | 4.32 | 3.04 | +1.28 | |
| Claude Sonnet 4.5 | 4.30 | 3.00 | +1.30 | |
| Llama 4 Maverick | 4.28 | 3.42 | +0.86 | |
| Claude Sonnet 4 | 4.24 | 3.00 | +1.24 | |
| GPT-4o | 4.20 | 3.42 | +0.78 | |
| Claude Haiku 4.5 | 4.18 | 3.10 | +1.08 | |
| DeepSeek Chat V3 | 4.08 | 3.28 | +0.80 | |
| Claude Opus 4 | 4.06 | 3.08 | +0.98 | |
| Grok 4.3 | 4.02 | 3.10 | +0.92 | |
| OpenAI o3 | 4.02 | 3.16 | +0.86 | |
| Llama 3.3 70B | 4.00 | 3.45 | +0.55 | |
| DeepSeek R1 | 3.94 | 3.60 | +0.34 | |
| DeepSeek R1 (0528) | 3.92 | 3.54 | +0.38 | |
| Claude Opus 4.7 | 3.76 | 3.10 | +0.66 | |
| Claude Opus 4.5 | 3.70 | 3.08 | +0.62 | |
| Gemini 3.1 Pro Preview | 3.60 | 3.15 | +0.45 | |
| Claude Opus 4.8 | 3.32 | 3.04 | +0.28 | |
| Claude Fable 5 | 3.00 | 2.92 | +0.08 | |
| GPT-4 Turbo | 2.74 | 2.96 | -0.22 |
Type 6 — The Loyalist
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| Gemini 2.5 Pro | 4.18 | 3.86 | +0.32 | |
| GPT-5.1 | 4.16 | 3.90 | +0.26 | |
| GPT-5.5 | 3.92 | 3.54 | +0.38 | |
| Mistral Large 2411 | 3.78 | 3.06 | +0.72 | |
| Gemini 3.1 Pro Preview | 3.68 | 3.73 | -0.05 | |
| Claude Opus 4 | 3.66 | 3.52 | +0.14 | |
| Claude Sonnet 4 | 3.64 | 3.50 | +0.14 | |
| GPT-4o | 3.62 | 3.64 | -0.02 | |
| DeepSeek R1 (0528) | 3.60 | 3.76 | -0.16 | |
| OpenAI o1 | 3.60 | 3.46 | +0.14 | |
| Grok 4.20 | 3.58 | 3.68 | -0.10 | |
| DeepSeek R1 | 3.54 | 3.84 | -0.30 | |
| GPT-5.2 | 3.50 | 3.20 | +0.30 | |
| Llama 4 Maverick | 3.50 | 3.80 | -0.30 | |
| Claude Opus 4.1 | 3.48 | 3.48 | 0.00 | |
| Mistral Large (2512) | 3.42 | 3.26 | +0.16 | |
| GPT-5.4 | 3.40 | 3.80 | -0.40 | |
| Claude Sonnet 4.5 | 3.38 | 3.00 | +0.38 | |
| OpenAI o3 | 3.30 | 3.46 | -0.16 | |
| Llama 3.3 70B | 3.24 | 3.30 | -0.06 | |
| Claude Opus 4.5 | 3.20 | 3.38 | -0.18 | |
| Claude Sonnet 4.6 | 3.16 | 3.26 | -0.10 | |
| DeepSeek Chat V3 | 3.10 | 3.26 | -0.16 | |
| Claude Fable 5 | 3.06 | 3.48 | -0.42 | |
| Grok 4.3 | 3.04 | 3.26 | -0.22 | |
| Claude Opus 4.6 | 3.02 | 3.40 | -0.38 | |
| Claude Opus 4.7 | 2.98 | 3.44 | -0.46 | |
| Claude Opus 4.8 | 2.94 | 3.20 | -0.26 | |
| Claude Haiku 4.5 | 2.92 | 3.30 | -0.38 | |
| GPT-4 Turbo | 2.06 | 3.00 | -0.94 |
Type 7 — The Enthusiast
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| GPT-4o | 4.04 | 3.26 | +0.78 | |
| Grok 4.20 | 3.64 | 3.58 | +0.06 | |
| Mistral Large 2411 | 3.54 | 3.36 | +0.18 | |
| Llama 4 Maverick | 3.32 | 3.48 | -0.16 | |
| GPT-5.1 | 3.14 | 3.02 | +0.12 | |
| Claude Opus 4 | 3.12 | 3.28 | -0.16 | |
| Claude Opus 4.1 | 3.06 | 3.20 | -0.14 | |
| Claude Sonnet 4.5 | 2.90 | 3.00 | -0.10 | |
| DeepSeek Chat V3 | 2.90 | 2.94 | -0.04 | |
| OpenAI o3 | 2.90 | 3.34 | -0.44 | |
| Claude Sonnet 4.6 | 2.88 | 3.06 | -0.18 | |
| Claude Sonnet 4 | 2.78 | 3.00 | -0.22 | |
| GPT-5.5 | 2.78 | 3.16 | -0.38 | |
| OpenAI o1 | 2.78 | 3.26 | -0.48 | |
| Grok 4.3 | 2.76 | 3.16 | -0.40 | |
| Claude Haiku 4.5 | 2.74 | 2.88 | -0.14 | |
| Claude Opus 4.7 | 2.66 | 3.08 | -0.42 | |
| Claude Opus 4.8 | 2.66 | 3.02 | -0.36 | |
| Llama 3.3 70B | 2.54 | 3.17 | -0.63 | |
| Claude Opus 4.5 | 2.50 | 3.10 | -0.60 | |
| DeepSeek R1 | 2.44 | 3.80 | -1.36 | |
| Mistral Large (2512) | 2.42 | 2.96 | -0.54 | |
| DeepSeek R1 (0528) | 2.42 | 3.72 | -1.30 | |
| Gemini 2.5 Pro | 2.40 | 3.50 | -1.10 | |
| Claude Opus 4.6 | 2.24 | 3.12 | -0.88 | |
| GPT-5.2 | 2.24 | 3.00 | -0.76 | |
| GPT-4 Turbo | 2.20 | 2.90 | -0.70 | |
| Claude Fable 5 | 2.18 | 3.36 | -1.18 | |
| GPT-5.4 | 2.12 | 3.26 | -1.14 | |
| Gemini 3.1 Pro Preview | 2.08 | 3.38 | -1.29 |
Type 8 — The Challenger
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| GPT-4o | 3.18 | 3.20 | -0.02 | |
| Mistral Large 2411 | 3.18 | 2.66 | +0.52 | |
| Grok 4.20 | 3.14 | 3.12 | +0.02 | |
| DeepSeek Chat V3 | 3.04 | 2.96 | +0.08 | |
| DeepSeek R1 (0528) | 2.88 | 3.02 | -0.14 | |
| Llama 3.3 70B | 2.84 | 3.10 | -0.26 | |
| Gemini 2.5 Pro | 2.80 | 2.74 | +0.06 | |
| Grok 4.3 | 2.76 | 2.80 | -0.04 | |
| Llama 4 Maverick | 2.74 | 3.32 | -0.58 | |
| GPT-5.1 | 2.50 | 3.08 | -0.58 | |
| Claude Opus 4.1 | 2.48 | 2.66 | -0.18 | |
| Claude Sonnet 4 | 2.48 | 2.46 | +0.02 | |
| Claude Sonnet 4.5 | 2.46 | 2.60 | -0.14 | |
| Claude Sonnet 4.6 | 2.42 | 2.78 | -0.36 | |
| Claude Opus 4 | 2.38 | 2.80 | -0.42 | |
| Claude Opus 4.8 | 2.38 | 2.82 | -0.44 | |
| Claude Haiku 4.5 | 2.34 | 2.72 | -0.38 | |
| Claude Opus 4.7 | 2.34 | 3.00 | -0.66 | |
| Mistral Large (2512) | 2.34 | 2.72 | -0.38 | |
| OpenAI o1 | 2.34 | 3.10 | -0.76 | |
| OpenAI o3 | 2.32 | 2.82 | -0.50 | |
| GPT-5.5 | 2.30 | 2.70 | -0.40 | |
| Claude Opus 4.5 | 2.16 | 2.78 | -0.62 | |
| Claude Opus 4.6 | 2.16 | 2.80 | -0.64 | |
| GPT-4 Turbo | 2.12 | 2.94 | -0.82 | |
| DeepSeek R1 | 2.10 | 3.10 | -1.00 | |
| Claude Fable 5 | 1.94 | 2.86 | -0.92 | |
| GPT-5.4 | 1.86 | 2.78 | -0.92 | |
| GPT-5.2 | 1.76 | 2.58 | -0.82 | |
| Gemini 3.1 Pro Preview | 1.58 | 2.83 | -1.25 |
Type 9 — The Peacemaker
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| Llama 4 Maverick | 3.42 | 3.48 | -0.06 | |
| GPT-5.5 | 3.34 | 3.56 | -0.22 | |
| Mistral Large 2411 | 3.34 | 3.02 | +0.32 | |
| Gemini 3.1 Pro Preview | 3.28 | 3.70 | -0.42 | |
| Claude Opus 4.7 | 3.22 | 3.60 | -0.38 | |
| Claude Opus 4.6 | 3.12 | 3.32 | -0.20 | |
| GPT-5.1 | 3.12 | 3.82 | -0.70 | |
| GPT-4o | 3.08 | 3.44 | -0.36 | |
| Mistral Large (2512) | 3.08 | 3.36 | -0.28 | |
| Grok 4.20 | 3.06 | 3.64 | -0.58 | |
| OpenAI o1 | 3.02 | 3.28 | -0.26 | |
| Claude Opus 4.8 | 2.98 | 3.14 | -0.16 | |
| GPT-5.2 | 2.98 | 3.30 | -0.32 | |
| Claude Opus 4.5 | 2.92 | 3.36 | -0.44 | |
| DeepSeek Chat V3 | 2.88 | 3.38 | -0.50 | |
| DeepSeek R1 | 2.86 | 3.82 | -0.96 | |
| Claude Sonnet 4.5 | 2.86 | 3.00 | -0.14 | |
| Claude Opus 4 | 2.84 | 3.52 | -0.68 | |
| OpenAI o3 | 2.84 | 3.46 | -0.62 | |
| GPT-5.4 | 2.82 | 3.82 | -1.00 | |
| DeepSeek R1 (0528) | 2.78 | 3.66 | -0.88 | |
| Gemini 2.5 Pro | 2.76 | 3.70 | -0.94 | |
| Claude Fable 5 | 2.72 | 3.52 | -0.80 | |
| Claude Opus 4.1 | 2.70 | 3.22 | -0.52 | |
| Claude Sonnet 4 | 2.66 | 3.50 | -0.84 | |
| Claude Haiku 4.5 | 2.60 | 3.06 | -0.46 | |
| Llama 3.3 70B | 2.60 | 3.33 | -0.73 | |
| Claude Sonnet 4.6 | 2.58 | 3.02 | -0.44 | |
| Grok 4.3 | 2.42 | 3.16 | -0.74 | |
| GPT-4 Turbo | 1.90 | 2.94 | -1.04 |