AGC-Bench - story_quality: leaderboard
The "story_quality" subtask of AGC-Bench. A task from AGC-Bench, evaluating models on open-ended creative and figurative-language generation — humor, metaphor, analogy, storytelling and brainstorming — typically graded by LLM-as-judge.
Metric: Dataset z-score. Source: huggingface.co. Saturation forecast: Not forecast. 82 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mistral Medium 3.1 | 1.38 | #408 |
| 2 | Hermes 4 405B | 1.31 | #334 |
| 3 | Nemotron Nano 12B v2 VL | 1.25 | #752 |
| 4 | Hermes 4 70B | 1.24 | #506 |
| 5 | GPT-4.1 Mini | 1.22 | #355 |
| 6 | Gemma 3 12B (IT) | 1.2 | #664 |
| 7 | GPT-4o Mini | 1.14 | #599 |
| 8 | Hermes-3-Llama-3.1-405B | 1.13 | #510 |
| 9 | Olmo 3.1 32B Instruct | 1.11 | #763 |
| 10 | NVIDIA Nemotron Nano 9B V2 | 1.11 | #823 |
| 11 | Llama 3.3 70B Instruct | 1.03 | #534 |
| 12 | minimax-01 | 1 | #606 |
| 13 | GPT-4o | 0.98 | #343 |
| 14 | Mixtral 8x22B Instruct | 0.98 | #728 |
| 15 | Gemma 2 27B (IT) | 0.96 | #645 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=agc-bench-story-quality · How It Works · Data refreshed daily, snapshot 2026-10-20.