BABILong (NIAH, v0, 10M): leaderboard
Metric: QA1–5 Mean Accuracy (%). Source: huggingface.co. Saturation forecast: Rough model projection: around 2026. 3 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | ~ ARMT (137M) fine-tune | 76.6 |
| 2 | Llama3-ChatQA-1.5-8B + RAG | 37 |
| 3 | ~ RMT (137M) fine-tune | 33.78 |
Interactive version: theaggregate.ai/benchmark?slug=babilong-niah-v0-10m · How It Works · Data refreshed daily, snapshot 2026-10-09.