HELM Long Context - InfiniteBench En.Sum: leaderboard
Metric: ROUGE-L. Source: crfm.stanford.edu. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 4 Scout Instruct | 17.6 |
| 2 | GPT-4.1 (2025-04-14) | 17.38 |
| 3 | Llama 4 Maverick Instruct FP8 | 16.06 |
| 4 | GPT-4.1 Mini | 16.05 |
| 5 | Gemini 2.0 Flash Lite | 15.51 |
| 6 | Gemini 2.0 Flash | 15.11 |
| 7 | Nova Lite | 14.76 |
| 8 | Palmyra X5 | 14.65 |
| 9 | Nova Premier | 13.72 |
| 10 | GPT-4.1 Nano | 11.34 |
| 11 | Nova Pro | 10.95 |
Interactive version: theaggregate.ai/benchmark?slug=helm-long-context-infinitebench-en-sum · How It Works · Data refreshed daily, snapshot 2026-09-05.