NarrativeWorldBench: leaderboard
Long-form serialized audio drama, with arcs that run for 200 to 800 episodes, is a major creative medium and a setting where frontier large language models (LLMs) fail. We benchmark 21 models, spanning classical, fine-tuned, open-frontier, closed-frontier, and reasoning tiers, on a uniform set of structural narrative metrics. All closed-frontier systems saturate at a plot-beat F1 in the band [0.78
Metric: Plot-Beat F1 (h=50) (self-reported). Source: benchmarklist.com. Saturation forecast: Around April 2027. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 81 | #98 |
| 2 | Claude Opus 4.5 (Thinking) | 80 | #83 (Claude Opus 4.5) |
| 3 | Gemini 2.5 Pro | 79 | #152 |
| 4 | O3 Pro | 79 | #116 |
| 5 | DeepSeek R1 | 78 | #251 |
| 6 | Llama 3.1 405B | 71 | #466 |
| 7 | GPT-3.5 Turbo | 54 | #865 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=narrativeworldbench · How It Works · Data refreshed daily, snapshot 2026-10-20.