LLM Stats (RULER 128k): leaderboard
LLM Stats relay of RULER at 128K context: synthetic retrieval, variable tracking, aggregation and question-answering tasks. Scores use the reported 0–100 scale; source-specific checkpoints and evaluation settings remain distinct.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around February 2027. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Granite 4.2 30B | 81.38 | #355 |
| 2 | Granite 4.2 8B | 71.41 | #525 |
| 3 | LFM3-1.7B | 65.56 | #501 |
| 4 | Granite 4.2 3B | 55.3 | #683 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-ruler-128k · How It Works · Data refreshed daily, snapshot 2026-11-05.