On Stable Long-Form Generation: leaderboard
Large Language Models (LLMs) excel at long-context understanding but exhibit significant limitations in long-form generation. Existing studies primarily focus on single-generation quality, generally overlooking the volatility of the output. This volatility not only leads to significant computational costs but also severely impacts the models' reliable application. To address this gap, our work unf
Metric: LVC (self-reported). Source: benchmarklist.com. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o Mini | 33.9 | #599 |
| 2 | Llama 3.1 8B | 26.5 | #1151 |
| 3 | Qwen 2.5 1.5B | 19.6 | #1431 |
| 4 | Qwen 2.5 7B | 17 | #922 |
| 5 | DeepSeek R1 | 8.6 | #251 |
| 6 | DeepSeek V3 | 2.2 | #321 |
| 7 | Claude 3.5 Sonnet | 1.9 | #350 |
Interactive version: theaggregate.ai/benchmark?slug=on-stable-long-form-generation · How It Works · Data refreshed daily, snapshot 2026-10-20.