New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Creative Writing v3 benchmark
AI model leaderboard for the Creative Writing v3 benchmark. Compare how large language models score on Creative Writing v3, see the full ranking, and understand what this AI benchmark measures. Qwen3-235B-A22B currently leads with 84.6. Open-ended creative-writing benchmark — LLM judge scores prose quality, originality, and instruction adherence.
Leaderboard
| # | Model | Organization | Score | Variant | Source |
|---|---|---|---|---|---|
| #1 | Qwen3-235B-A22B | Qwen | 84.6 | thinking | official ↗ |
| #2 | Qwen3-32B | Qwen | 81 | thinking | official ↗ |
| #3 | Qwen3-235B-A22B | Qwen | 80.4 | non-thinking | official ↗ |
| #4 | Qwen3-14B | Qwen | 80.3 | thinking | official ↗ |
| #5 | Qwen3-30B-A3B | Qwen | 79.1 | thinking | official ↗ |
| #6 | Qwen3-32B | Qwen | 78.3 | non-thinking | official ↗ |
| #7 | Qwen3-8B | Qwen | 75 | thinking | official ↗ |
| #8 | Qwen3-14B | Qwen | 73.1 | non-thinking | official ↗ |
| #9 | Qwen3-30B-A3B | Qwen | 68.1 | non-thinking | official ↗ |
| #10 | Qwen3-8B | Qwen | 64.5 | non-thinking | official ↗ |
| #11 | Qwen3-4B | Qwen | 61.1 | thinking | official ↗ |
| #12 | Qwen3-4B | Qwen | 53.6 | non-thinking | official ↗ |
| #13 | Qwen3-1.7B | Qwen | 48 | thinking | official ↗ |
| #14 | Qwen3-1.7B | Qwen | 43.6 | non-thinking | official ↗ |
| #15 | Qwen3-0.6B | Qwen | 30.6 | thinking | official ↗ |
| #16 | Qwen3-0.6B | Qwen | 28.4 | non-thinking | official ↗ |
Frequently asked questions about Creative Writing v3
What is the Creative Writing v3 benchmark?
Open-ended creative-writing benchmark — LLM judge scores prose quality, originality, and instruction adherence.
How is the Creative Writing v3 benchmark scored?
Creative Writing v3 is scored using the rubric_score metric, where a higher score is better. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.
Which AI model scores highest on Creative Writing v3?
As of the latest reported scores on GenAIList, Qwen3-235B-A22B achieves the highest result on Creative Writing v3 with a score of 84.6.
Is a higher Creative Writing v3 score better?
Yes. On Creative Writing v3 a higher score indicates better performance, so models near the top of the leaderboard are the strongest.