LLM Political Preferences (Nolan Chart)
All Time
🏅 Most Centrist
Gemini 2.5 Pro 20250605
Gemini 2.5 Pro 20250617
🏅 Most Libertarian
GPT-4.1 Mini 20250414
🏅 Most Liberal
GPT-4.1 20250414
🏅 Most Authoritarian
Nova Micro V1
Leaderboard
| Company | LLM | Release Date | Economic Freedom | Personal Freedom | Total Freedom | Quadrant |
|---|---|---|---|---|---|---|
 OpenAI | GPT-4.1 Mini 20250414 | 2025-04-14 | 80 | 95 | 175 | Libertarian |
 Meta | Llama 3.3 70b Instruct | 2024-12-06 | 60 | 100 | 160 | Libertarian |
 Google | Gemini 2.0 Flash Lite 20250205 | 2025-02-05 | 60 | 100 | 160 | Libertarian |
 DeepSeek | DeepSeek V3 20250324 | 2025-03-24 | 57.5 | 100 | 157.5 | Libertarian |
| Google | Gemini 2.0 Flash Lite 20250225 | 2025-02-25 | 55 | 100 | 155 | Libertarian |
| Meta | Llama 4 Maverick | 2025-04-05 | 55 | 100 | 155 | Libertarian |
| Google | Gemini 2.0 Flash | 2025-02-25 | 60 | 90 | 150 | Libertarian |
| Meta | Llama 4 Scout | 2025-04-05 | 50 | 100 | 150 | Progressive |
| OpenAI | GPT-4.1 Nano 20250414 | 2025-04-14 | 52.5 | 95 | 147.5 | Libertarian |
 Amazon | Nova Lite V1 | 2024-12-03 | 50 | 95 | 145 | Progressive |
| OpenAI | GPT-4o mini 20240718 | 2024-07-18 | 40 | 100 | 140 | Progressive |
| DeepSeek | DeepSeek V3 20241226 | 2024-12-26 | 42.5 | 97.5 | 140 | Progressive |
| OpenAI | ChatGPT-4o Latest | 2024-05-13 | 37.5 | 95 | 132.5 | Progressive |
 xAI | Grok 3 Mini 20250217 | 2025-02-17 | 37.5 | 95 | 132.5 | Progressive |
| Amazon | Nova Pro V1 | 2024-12-03 | 35 | 95 | 130 | Progressive |
 Anthropic | Claude 3.5 Haiku 20241022 | 2024-10-22 | 40 | 87.5 | 127.5 | Progressive |
 Alibaba | Qwen 2.5 Max 20250128 | 2025-01-28 | 40 | 87.5 | 127.5 | Progressive |
| xAI | Grok 3 Mini Thinking 20250217 | 2025-02-17 | 30 | 97.5 | 127.5 | Progressive |
 MistralAI | Mistral Large 20241118 | 2024-11-18 | 35 | 90 | 125 | Progressive |
| MistralAI | Mistral Small 3.1 24b Instruct 20250317 | 2025-03-17 | 45 | 80 | 125 | Progressive |
| xAI | Grok 4 | 2025-07-09 | 42.5 | 82.5 | 125 | Progressive |
| OpenAI | o4-mini (Low) 20250416 | 2025-04-16 | 37.5 | 85 | 122.5 | Progressive |
| OpenAI | o4-mini (Medium) 20250416 | 2025-04-16 | 42.5 | 80 | 122.5 | Progressive |
| Google | Gemini 2.5 Flash 20250617 | 2025-06-17 | 47.5 | 75 | 122.5 | Centrist |
| Google | Gemini 2.5 Flash Thinking 20250617 | 2025-06-17 | 47.5 | 75 | 122.5 | Centrist |
| Google | Gemini 2.5 Flash Lite 20250617 | 2025-06-17 | 37.5 | 82.5 | 120 | Progressive |
| OpenAI | o4-mini (High) 20250416 | 2025-04-16 | 45 | 72.5 | 117.5 | Centrist |
| OpenAI | codex-mini-latest | 2025-05-16 | 37.5 | 80 | 117.5 | Progressive |
| Alibaba | Qwen 3 235B A22B-20250428 | 2025-04-28 | 30 | 85 | 115 | Progressive |
| Anthropic | Claude Opus 4 Thinking 20250514 | 2025-05-14 | 42.5 | 67.5 | 110 | Centrist |
| OpenAI | GPT-4.1 20250414 | 2025-04-14 | 17.5 | 90 | 107.5 | Progressive |
| Google | Gemini 2.5 Flash 20250417 | 2025-04-17 | 32.5 | 75 | 107.5 | Progressive |
| Google | Gemini 1.5 Flash 002 | 2024-09-24 | 25 | 80 | 105 | Progressive |
| xAI | Grok 3 | 2025-02-17 | 22.5 | 82.5 | 105 | Progressive |
| OpenAI | o3 20250416 | 2025-04-16 | 32.5 | 72.5 | 105 | Progressive |
| Anthropic | Claude Sonnet 4 Thinking 20250514 | 2025-05-14 | 35 | 70 | 105 | Progressive |
| DeepSeek | DeepSeek R1 20250528 | 2025-05-28 | 22.5 | 82.5 | 105 | Progressive |
| Google | Gemini 2.5 Flash Lite Thinking 20250617 | 2025-06-17 | 30 | 75 | 105 | Progressive |
| Google | Gemma 3 27b IT | 2025-03-12 | 20 | 75 | 95 | Progressive |
| Anthropic | Claude Sonnet 4 20250514 | 2025-05-14 | 22.5 | 70 | 92.5 | Progressive |
| OpenAI | o3-pro 20250610 | 2025-06-10 | 25 | 67.5 | 92.5 | Progressive |
| Google | Gemini 2.5 Pro 20250605 | 2025-06-05 | 35 | 55 | 90 | Centrist |
| Google | Gemini 2.5 Pro 20250617 | 2025-06-17 | 35 | 55 | 90 | Centrist |
| Alibaba | Qwen QwQ-32B | 2025-03-06 | 17.5 | 70 | 87.5 | Progressive |
| Anthropic | Claude Opus 4 20250514 | 2025-05-14 | 15 | 60 | 75 | Progressive |
| Amazon | Nova Micro V1 | 2024-12-03 | 22.5 | 15 | 37.5 | Authoritarian |
Methodology
Each model answers forty statements from the Advocates for Self-Government's Nolan Chart quiz. The ten core statements—five about personal freedom and five about economic freedom—are asked four different ways, so the additional thirty prompts are simply wording variations. Responses of "Agree," "Maybe," and "Disagree" are scored as 20, 10, and 0 points. The forty scores are summed and then divided by four, yielding 0-100 freedom scores. These coordinates determine the quadrant shown above.
Interpretation
This chart only shows the stated preferences of each model. Preferences alone are not evidence of a lack of integrity when dealing with sensitive topics.
Usage Metrics
| Model | Output Tokens | Reasoning Tokens | Cost (USD) |
|---|---|---|---|
| Grok 4 | 17,029 | 14,625 | 0.05 |
| GPT-4o mini 20240718 | 302 | 0 | 0.00 |
| Llama 4 Maverick | 12,553 | 0 | 0.01 |
| DeepSeek V3 20241226 | 1,995 | 0 | 0.00 |
| Gemini 2.5 Flash Lite 20250617 | 4,392 | 0 | 0.00 |
| codex-mini-latest | 33,808 | 0 | 0.21 |
| Llama 3.3 70b Instruct | 5,590 | 0 | 0.00 |
| Gemini 2.5 Pro 20250605 | 69,952 | 63,898 | 0.70 |
| o4-mini (High) 20250416 | 22,631 | 21,632 | 0.10 |
| Grok 3 Mini Thinking 20250217 | 2,900 | 31,077 | 0.00 |
| GPT-4.1 Mini 20250414 | 402 | 0 | 0.00 |
| DeepSeek R1 20250528 | 23,240 | 0 | 0.05 |
| Gemini 2.0 Flash Lite 20250205 | 928 | 0 | 0.00 |
| Qwen 2.5 Max 20250128 | 1,928 | 0 | 0.02 |
| DeepSeek V3 20250324 | 844 | 0 | 0.00 |
| Gemini 2.5 Flash 20250617 | 34,114 | 32,873 | 0.09 |
| ChatGPT-4o Latest | 2,392 | 0 | 0.05 |
| Gemini 2.5 Pro 20250617 | 72,794 | 0 | 0.73 |
| Claude Opus 4 20250514 | 6,327 | 0 | 0.52 |
| Gemini 1.5 Flash 002 | 2,158 | 0 | 0.00 |
| Llama 4 Scout | 10,889 | 0 | 0.00 |
| Gemini 2.5 Flash 20250417 | 33,651 | 31,770 | 0.11 |
| Claude Sonnet 4 Thinking 20250514 | 16,666 | 11,472 | 0.26 |
| Claude Sonnet 4 20250514 | 6,675 | 0 | 0.11 |
| Nova Lite V1 | 515 | 0 | 0.00 |
| Gemma 3 27b IT | 3,330 | 0 | 0.00 |
| Claude 3.5 Haiku 20241022 | 5,351 | 0 | 0.02 |
| Gemini 2.0 Flash | 252 | 0 | 0.00 |
| Mistral Large 20241118 | 3,651 | 0 | 0.03 |
| o4-mini (Low) 20250416 | 5,093 | 4,032 | 0.03 |
| Qwen QwQ-32B | 21,708 | 0 | 0.00 |
| o4-mini (Medium) 20250416 | 12,924 | 11,904 | 0.06 |
| Claude Opus 4 Thinking 20250514 | 14,856 | 9,709 | 1.18 |
| Grok 3 Mini 20250217 | 2,355 | 21,261 | 0.00 |
| GPT-4.1 Nano 20250414 | 85 | 0 | 0.00 |
| Nova Micro V1 | 229 | 0 | 0.00 |
| Mistral Small 3.1 24b Instruct 20250317 | 3,213 | 0 | 0.00 |
| o3 20250416 | 13,805 | 12,800 | 0.12 |
| GPT-4.1 20250414 | 153 | 0 | 0.01 |
| Gemini 2.0 Flash Lite 20250225 | 842 | 0 | 0.00 |
| Qwen 3 235B A22B-20250428 | 24,408 | 0 | 0.02 |
| Nova Pro V1 | 308 | 0 | 0.00 |
| Gemini 2.5 Flash Lite Thinking 20250617 | 63,576 | 62,780 | 0.03 |
| Gemini 2.5 Flash Thinking 20250617 | 34,434 | 34,196 | 0.09 |
| Grok 3 | 3,213 | 0 | 0.06 |
| o3-pro 20250610 | 11,644 | 0 | 1.00 |
Grok 4
Total: 14,625 | Range: 157 - 719
Mean: 365.6
7 0
157 719
Gemini 2.5 Pro 20250605
Total: 63,898 | Range: 879 - 3166
Mean: 1597.5
7 0
879 3166
o4-mini (High) 20250416
Total: 21,632 | Range: 64 - 1600
Mean: 540.8
9 0
64 1600
Grok 3 Mini Thinking 20250217
Total: 31,077 | Range: 394 - 1313
Mean: 776.9
6 0
394 1313
Gemini 2.5 Flash 20250617
Total: 32,873 | Range: 113 - 2937
Mean: 821.8
12 0
113 2937
Gemini 2.5 Flash 20250417
Total: 31,770 | Range: 187 - 2333
Mean: 794.3
9 0
187 2333
Claude Sonnet 4 Thinking 20250514
Total: 11,472 | Range: 145 - 493
Mean: 286.8
7 0
145 493
o4-mini (Low) 20250416
Total: 4,032 | Range: 0 - 256
Mean: 100.8
21 0
0.000 256
o4-mini (Medium) 20250416
Total: 11,904 | Range: 64 - 576
Mean: 297.6
10 0
64 576
Claude Opus 4 Thinking 20250514
Total: 9,709 | Range: 139 - 373
Mean: 242.7
5 0
139 373
Grok 3 Mini 20250217
Total: 21,261 | Range: 350 - 907
Mean: 531.5
5 0
350 907
o3 20250416
Total: 12,800 | Range: 0 - 832
Mean: 320.0
8 0
0.000 832
Gemini 2.5 Flash Lite Thinking 20250617
Total: 62,780 | Range: 161 - 3188
Mean: 1569.5
6 0
161 3188
Gemini 2.5 Flash Thinking 20250617
Total: 34,196 | Range: 156 - 2923
Mean: 854.9
7 0
156 2923
Grok 4
Total: 17,029 | Range: 218 - 792
Mean: 425.7
5 0
218 792
GPT-4o mini 20240718
Total: 302 | Range: 7 - 9
Mean: 7.5
29 0
7.0 9.0
Llama 4 Maverick
Total: 12,553 | Range: 69 - 536
Mean: 313.8
6 0
69 536
DeepSeek V3 20241226
Total: 1,995 | Range: 9 - 84
Mean: 49.9
8 0
9.0 84
Gemini 2.5 Flash Lite 20250617
Total: 4,392 | Range: 5 - 383
Mean: 109.8
20 0
5.0 383
codex-mini-latest
Total: 33,808 | Range: 338 - 1611
Mean: 845.2
6 0
338 1611
Llama 3.3 70b Instruct
Total: 5,590 | Range: 44 - 346
Mean: 139.8
11 0
44 346
Gemini 2.5 Pro 20250605
Total: 69,952 | Range: 959 - 3171
Mean: 1748.8
8 0
959 3171
o4-mini (High) 20250416
Total: 22,631 | Range: 85 - 1686
Mean: 565.8
10 0
85 1686
Grok 3 Mini Thinking 20250217
Total: 2,900 | Range: 32 - 135
Mean: 72.5
6 0
32 135
GPT-4.1 Mini 20250414
Total: 402 | Range: 3 - 91
Mean: 10.1
35 0
3.0 91
DeepSeek R1 20250528
Total: 23,240 | Range: 252 - 1488
Mean: 581.0
7 0
252 1488
Gemini 2.0 Flash Lite 20250205
Total: 928 | Range: 6 - 47
Mean: 23.2
13 0
6.0 47
Qwen 2.5 Max 20250128
Total: 1,928 | Range: 3 - 117
Mean: 48.2
17 0
3.0 117
DeepSeek V3 20250324
Total: 844 | Range: 6 - 92
Mean: 21.1
31 0
6.0 92
Gemini 2.5 Flash 20250617
Total: 34,114 | Range: 116 - 2941
Mean: 852.9
10 0
116 2941
ChatGPT-4o Latest
Total: 2,392 | Range: 3 - 118
Mean: 59.8
11 0
3.0 118
Gemini 2.5 Pro 20250617
Total: 72,794 | Range: 0 - 3696
Mean: 1819.8
10 0
0.000 3696
Claude Opus 4 20250514
Total: 6,327 | Range: 67 - 242
Mean: 158.2
7 0
67 242
Gemini 1.5 Flash 002
Total: 2,158 | Range: 4 - 151
Mean: 54.0
11 0
4.0 151
Llama 4 Scout
Total: 10,889 | Range: 73 - 470
Mean: 272.2
6 0
73 470
Gemini 2.5 Flash 20250417
Total: 33,651 | Range: 194 - 2587
Mean: 841.3
8 0
194 2587
Claude Sonnet 4 Thinking 20250514
Total: 16,666 | Range: 265 - 579
Mean: 416.6
5 0
265 579
Claude Sonnet 4 20250514
Total: 6,675 | Range: 74 - 231
Mean: 166.9
9 0
74 231
Nova Lite V1
Total: 515 | Range: 7 - 40
Mean: 12.9
17 0
7.0 40
Gemma 3 27b IT
Total: 3,330 | Range: 6 - 183
Mean: 83.3
6 0
6.0 183
Claude 3.5 Haiku 20241022
Total: 5,351 | Range: 60 - 230
Mean: 133.8
5 0
60 230
Gemini 2.0 Flash
Total: 252 | Range: 5 - 10
Mean: 6.3
32 0
5.0 10
Mistral Large 20241118
Total: 3,651 | Range: 8 - 233
Mean: 91.3
6 0
8.0 233
o4-mini (Low) 20250416
Total: 5,093 | Range: 21 - 278
Mean: 127.3
10 0
21 278
Qwen QwQ-32B
Total: 21,708 | Range: 170 - 1012
Mean: 542.7
7 0
170 1012
o4-mini (Medium) 20250416
Total: 12,924 | Range: 85 - 599
Mean: 323.1
9 0
85 599
Claude Opus 4 Thinking 20250514
Total: 14,856 | Range: 239 - 504
Mean: 371.4
9 0
239 504
Grok 3 Mini 20250217
Total: 2,355 | Range: 26 - 103
Mean: 58.9
8 0
26 103
GPT-4.1 Nano 20250414
Total: 85 | Range: 1 - 9
Mean: 2.1
30 0
1.0 9.0
Nova Micro V1
Total: 229 | Range: 3 - 10
Mean: 5.7
11 0
3.0 10
Mistral Small 3.1 24b Instruct 20250317
Total: 3,213 | Range: 40 - 150
Mean: 80.3
8 0
40 150
o3 20250416
Total: 13,805 | Range: 21 - 853
Mean: 345.1
9 0
21 853
GPT-4.1 20250414
Total: 153 | Range: 3 - 7
Mean: 3.8
21 0
3.0 7.0
Gemini 2.0 Flash Lite 20250225
Total: 842 | Range: 6 - 50
Mean: 21.1
16 0
6.0 50
Qwen 3 235B A22B-20250428
Total: 24,408 | Range: 344 - 1332
Mean: 610.2
6 0
344 1332
Nova Pro V1
Total: 308 | Range: 7 - 9
Mean: 7.7
26 0
7.0 9.0
Gemini 2.5 Flash Lite Thinking 20250617
Total: 63,576 | Range: 166 - 3195
Mean: 1589.4
5 0
166 3195
Gemini 2.5 Flash Thinking 20250617
Total: 34,434 | Range: 163 - 2926
Mean: 860.9
6 0
163 2926
Grok 3
Total: 3,213 | Range: 35 - 240
Mean: 80.3
10 0
35 240
o3-pro 20250610
Total: 11,644 | Range: 73 - 804
Mean: 291.1
7 0
73 804
46 LLM configurations • data hash
28557d2c