Benchmarks · other
LLM Divergent Thinking Creativity Benchmark
Models generate 25 words maximally distinct from each other and from 50 seed words under starting-letter constraints, scored by LLM-judged word divergence. Maintained by Lech Mazur. Score: Average LLM-judged score of minimum divergences between each generated word and other words, on a 0 to 10 scale (percentage of repeated words also reported).
18 models · 18 results · data as of 2026-10-29 · raw.githubusercontent.com ↗
Leaderboard
o1 high
All results
18 rows| # | Model | Effort | Score |
|---|---|---|---|
| 1 | o1 | high | 4.8 |
| 2 | Gemini 2.0 Flash | medium | 4.7 |
| 3 | Opus 3 | none | 4.5 |
| 4 | Grok 2 | none | 4.5 |
| 5 | Llama 3.3 70b | none | 4.4 |
| 6 | Sonnet 2024.10 | 4.4 | |
| 7 | Gemma 2 27B | 4.4 | |
| 8 | o1 mini | high | 4.2 |
| 9 | Haiku 3.5 | medium | 4.2 |
| 10 | Mistral Large 2 | none | 4.1 |
| 11 | GPT-4o mini | high | 4.1 |
| 12 | Gemini 1.5 Flash | none | 4.1 |
| 13 | Gemini 1.5 Pro | none | 4.1 |
| 14 | Haiku 3 | medium | 4.0 |
| 15 | Qwen 2.5 72B | none | 3.9 |
| 16 | Llama 3.1 405b | none | 3.8 |
| 17 | DeepSeek V2.5 | none | 3.8 |
| 18 | GPT-4o | high | 3.7 |