FMNB Leaderboard: leaderboard
Fiction Multi-Needle Benchmark: long-context evaluation testing models on retrieving and synthesizing information from multiple locations in fiction texts at 8K-32K context lengths.
Metric: Score. Source: huggingface.co. Saturation forecast: Estimated already saturated. 49 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3.3 70B Instruct | 100 | #534 |
| 2 | Gemma 4 31B (IT) | 100 | #176 |
| 3 | Qwen 3.5 27B | 100 | #187 |
| 4 | GLM 4.5 Air | 99 | #392 |
| 5 | Mistral Small 3.2 | 93 | #605 |
| 6 | Magistral Small 1.2 | 88 | #570 |
| 7 | Llama 4 Scout Instruct | 83 | #539 |
| 8 | Mistral Nemo Instruct (2407) | 55 | #936 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=fmnb-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.