Ruler 128k leaderboard
As of November 14, 2026, among the results listed here, IBM Granite 4.2 8B has the highest reported Ruler 128k score, 71.4, published by IBM Granite, ahead of LiquidAI LFM3-1.7B at 65.6. 3 models have a Ruler 128k result in our records.
Scores as the model makers published them, not re-run by us; each row links to its announcement.
3 models ranked · 2 makers
| Rank | Model | Score | Setting | Reported by | Date |
|---|---|---|---|---|---|
| 1 | Granite 4.2 8B | 71.4 | Not published | IBM Granite | August 31, 2026 (model release) |
| 2 | LFM3-1.7B | 65.6 | Not published | Liquid | November 6, 2026 (model release) |
| 3 | LFM2.5-2.6BRetired | 1.2 | Not published | Liquid | November 6, 2026 (model release) |
How much to trust these numbers
- Source
- Reported by the model makers themselves. Not re-run by LLMBenchmark or anyone independent.
- Settings
- All ranked results use the same reported setting.
- Spread
- 71.4 to 1.2 across 3 models.
- Freshness
- Newest result published November 6, 2026; oldest August 31, 2026.
What Ruler 128k measures
RULER 128k tests models on 13 long-context tasks, including retrieval, within 131,072 tokens; its score indicates performance on those tasks.