RULER (1M)
RULER (1M) - long context retrieval benchmark at 1M tokens.
Category
Long Context
Max Score
100.0
Score Type
percent
Active
Yes
Model Scores
Scores for RULER (1M)
| Model | Score | Date | Verified |
|---|---|---|---|
| NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 |
100.00%
|
04.06.2026 | Verified |
| Qwen3.5-397B-A17B |
34.29%
|
04.06.2026 | Unverified |
| DeepSeek-V4-Pro |
92.86%
|
04.06.2026 | Unverified |
| DeepSeek-V4-Flash |
87.70
|
04.06.2026 | Unverified |