BABILong (NIAH, v0, 64k): leaderboard
Metric: QA1–5 Mean Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.1 70B Instruct | 53 |
| 2 | Llama 3.1 8B Instruct | 49 |
| 3 | Yi-34B-200K | 48 |
| 4 | c4ai-command-r-v01 | 46 |
| 5 | Phi-3-medium-128k-instruct | 45 |
| 6 | GPT-4 Preview (0125) | 43 |
| 7 | Phi-3 Mini 128K Instruct | 37 |
| 8 | Mixtral 8x22B Instruct (v0.1) | 35 |
| 9 | Yi-9B-200K | 29 |
| 10 | Yarn-Mistral-7B-128k | 10 |
Interactive version: theaggregate.ai/benchmark?slug=babilong-niah-v0-64k · How It Works · Data refreshed daily, snapshot 2026-10-09.