Head to head
Laya Router scores 60.4 and d3 nano 78.0 on the System One Index. Laya Router is more accurate on 2 of 16 text datasets, d3 nano on 13, and 1 are tied. Same cases, same request, same scorer.
| Laya Router | d3 nano | |
|---|---|---|
| System One Index | 60.4 | 78.0 |
| English | 64.7 | 77.9 |
| Arabic | 53.2 | 76.9 |
| Vision Index | – | – |
| Calibration error | 20.0% | 7.0% |
| Median latency | – | – |
| Cost per 1,000 calls | – | – |
| Laya Router | d3 nano | |
|---|---|---|
| AG News English | 92.5 | 88.0 |
| SST-2 English | 88.5 | 89.5 |
| SST-5 English | 34.0 | 55.5 |
| DAIR Emotion English | 53.5 | 59.5 |
| Banking77 English | 44.0 | 93.0 |
| BoolQ English | 81.5 | 88.0 |
| XNLI English English | 87.3 | 87.3 |
| Typed decisions English | 36.4 | 62.6 |
| XNLI French French | 76.0 | 86.0 |
| XNLI Arabic Arabic | 68.0 | 82.0 |
| MASSIVE scenario Arabic | 39.0 | 61.5 |
| MASSIVE intent Arabic | 32.5 | 81.0 |
| OSACT4 offensive Arabic | 82.5 | 79.0 |
| Tweet sentiment Arabic | 37.0 | 71.0 |
| MIRACL rerank Arabic | 31.5 | 77.5 |
| MIRACL pairwise Arabic | 82.0 | 86.5 |
Model pages: Laya Router · d3 nano. Latency for API models is the median per call through OpenRouter, timed from one machine.