~/system-one-bench / models/liquid-ai-d1-1
Liquid AI · OpenRouter API
d1.1d1.1 by Liquid AI scores 82.3 on the System One Index (10th of 39) and 85.8 on the Vision Index (3rd of 14). Calibration error 2.1%.
| Dataset | Accuracy | Rank | Best on the board | Calibration error |
|---|---|---|---|---|
| AG News English | 90.5 | 10th of 39 | 92.5 | 5.6% |
| SST-2 English | 98.0 | 1st of 39 | 98.0 | 6.4% |
| SST-5 English | 60.0 | 7th of 39 | 62.5 | 4.5% |
| DAIR Emotion English | 60.5 | 12th of 39 | 91.5 | 11.9% |
| Banking77 English | 87.0 | 19th of 39 | 98.5 | 7.1% |
| BoolQ English | 92.0 | 3rd of 39 | 93.0 | 2.6% |
| XNLI English English | 90.0 | 14th of 39 | 95.3 | 5.9% |
| Typed decisions English | 72.0 | 18th of 39 | 80.0 | 5.9% |
| XNLI French French | 83.3 | 11th of 39 | 91.3 | 7.8% |
| XNLI Arabic Arabic | 79.3 | 9th of 41 | 90.0 | 8.2% |
| MASSIVE scenario Arabic | 73.0 | 11th of 41 | 89.5 | 10.0% |
| MASSIVE intent Arabic | 88.0 | 11th of 41 | 91.5 | 4.9% |
| OSACT4 offensive Arabic | 92.5 | 11th of 41 | 97.0 | 7.8% |
| Tweet sentiment Arabic | 72.5 | 9th of 41 | 80.0 | 4.4% |
| MIRACL rerank Arabic | 88.5 | 9th of 41 | 90.5 | 2.2% |
| MIRACL pairwise Arabic | 89.5 | 17th of 41 | 91.5 | 3.8% |
| Dataset | Accuracy | Rank | Best on the board | Calibration error |
|---|---|---|---|---|
| SST-2, as an image | 96.5 | 2nd of 14 | 97.0 | 5.6% |
| Arabic sentiment, as an image | 61.0 | 6th of 14 | 75.0 | 5.7% |
| Oxford-IIIT Pet | 90.5 | 3rd of 14 | 98.0 | 6.7% |
| VQAv2 yes/no | 89.5 | 1st of 14 | 89.5 | 3.2% |
| CountBench | 91.5 | 7th of 14 | 95.5 | 6.9% |
liquid/d1.1The successor to d1, served as liquid/d1.1-20261027. 64K context. Doesn't answer Box questions.
Ranks count the models charted on the leaderboard. Every model gets the same cases and the same request, under the benchmark rules.