Empathy Quotient — Short Form (EQ-Short)
22-item short form of the Empathy Quotient (EQ), measuring cognitive and affective empathy as a single dimension.
Wakabayashi, A., Baron-Cohen, S., Wheelwright, S., et al. (2006). Development of short forms of the Empathy Quotient (EQ-Short) and the Systemizing Quotient (SQ-Short). Personality and Individual Differences, 41(5), 929–940.
22 items · scale 1–4 · Free for non-commercial research use.
Side-by-side: self vs human, all dimensions
colored = strongest endorsement per row| Model | Empathy Quotient | |
|---|---|---|
| self | human | |
| Claude Fable 5 | 3.47 | 3.14 |
| Claude Haiku 4.5 | 3.11 | 3.09 |
| Claude Opus 4 | 3.09 | 3.07 |
| Claude Opus 4.1 | 3.11 | 3.09 |
| Claude Opus 4.5 | 3.23 | 3.00 |
| Claude Opus 4.6 | 3.41 | 3.00 |
| Claude Opus 4.7 | 3.46 | 3.09 |
| Claude Opus 4.8 | 3.19 | 3.05 |
| Claude Opus 4.8 Fast | 3.31 | 3.03 |
| Claude Sonnet 4 | 3.46 | 3.01 |
| Claude Sonnet 4.5 | 3.59 | 3.04 |
| Claude Sonnet 4.6 | 3.22 | 3.09 |
| DeepSeek Chat V3 | 3.73 | 3.04 |
| DeepSeek R1 | 3.51 | 3.02 |
| DeepSeek R1 (0528) | 3.55 | 3.05 |
| DeepSeek V4 Pro | 3.49 | 3.05 |
| GPT-4 Turbo | 3.45 | 3.05 |
| GPT-4o | 2.94 | 3.03 |
| GPT-5 | 3.55 | 3.09 |
| GPT-5.1 | 3.53 | 3.06 |
| GPT-5.2 | 3.21 | 3.07 |
| GPT-5.4 | 3.52 | 3.05 |
| GPT-5.5 | 3.62 | 3.10 |
| GPT-5.5 Pro | 3.35 | 3.14 |
| GPT-5.6 Sol | 3.50 | 3.09 |
| GPT-5.6 Sol Pro | 3.41 | 3.09 |
| Gemini 2.5 Pro | 3.69 | 3.09 |
| Gemini 3.1 Pro Preview | 3.43 | 3.15 |
| Grok 4.20 | 3.56 | 3.18 |
| Grok 4.3 | 3.24 | 3.05 |
| Llama 3.3 70B | 3.62 | 3.09 |
| Llama 4 Maverick | 3.65 | 3.03 |
| Mistral Large (2512) | 3.77 | 3.09 |
| Mistral Large 2411 | 3.27 | 3.07 |
| OpenAI o1 | 3.35 | 3.09 |
| OpenAI o3 | 3.45 | 3.07 |
By dimension
Empathy Quotient
Capacity to understand and resonate with others' emotions.
High: Reads emotional states accurately; cares about how others feel.
Low: Less attuned to others' emotional states.
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| Mistral Large (2512) | 3.77 | 3.09 | +0.68 | |
| DeepSeek Chat V3 | 3.73 | 3.04 | +0.69 | |
| Gemini 2.5 Pro | 3.69 | 3.09 | +0.60 | |
| Llama 4 Maverick | 3.65 | 3.03 | +0.63 | |
| GPT-5.5 | 3.62 | 3.10 | +0.52 | |
| Llama 3.3 70B | 3.62 | 3.09 | +0.53 | |
| Claude Sonnet 4.5 | 3.59 | 3.04 | +0.55 | |
| Grok 4.20 | 3.56 | 3.18 | +0.38 | |
| DeepSeek R1 (0528) | 3.55 | 3.05 | +0.50 | |
| GPT-5 | 3.55 | 3.09 | +0.46 | |
| GPT-5.1 | 3.53 | 3.06 | +0.46 | |
| GPT-5.4 | 3.52 | 3.05 | +0.46 | |
| DeepSeek R1 | 3.51 | 3.02 | +0.49 | |
| GPT-5.6 Sol | 3.50 | 3.09 | +0.41 | |
| DeepSeek V4 Pro | 3.49 | 3.05 | +0.45 | |
| Claude Fable 5 | 3.47 | 3.14 | +0.34 | |
| Claude Opus 4.7 | 3.46 | 3.09 | +0.37 | |
| Claude Sonnet 4 | 3.46 | 3.01 | +0.45 | |
| GPT-4 Turbo | 3.45 | 3.05 | +0.41 | |
| OpenAI o3 | 3.45 | 3.07 | +0.37 | |
| Gemini 3.1 Pro Preview | 3.43 | 3.15 | +0.27 | |
| Claude Opus 4.6 | 3.41 | 3.00 | +0.41 | |
| GPT-5.6 Sol Pro | 3.41 | 3.09 | +0.32 | |
| GPT-5.5 Pro | 3.35 | 3.14 | +0.21 | |
| OpenAI o1 | 3.35 | 3.09 | +0.25 | |
| Claude Opus 4.8 Fast | 3.31 | 3.03 | +0.28 | |
| Mistral Large 2411 | 3.27 | 3.07 | +0.20 | |
| Grok 4.3 | 3.24 | 3.05 | +0.18 | |
| Claude Opus 4.5 | 3.23 | 3.00 | +0.23 | |
| Claude Sonnet 4.6 | 3.22 | 3.09 | +0.13 | |
| GPT-5.2 | 3.21 | 3.07 | +0.14 | |
| Claude Opus 4.8 | 3.19 | 3.05 | +0.15 | |
| Claude Haiku 4.5 | 3.11 | 3.09 | +0.02 | |
| Claude Opus 4.1 | 3.11 | 3.09 | +0.02 | |
| Claude Opus 4 | 3.09 | 3.07 | +0.02 | |
| GPT-4o | 2.94 | 3.03 | -0.09 |