Head to head
Microsoft-Decision-1 scores 82.0 and Solar Decide Flash 63.6 on the System One Index. Microsoft-Decision-1 is more accurate on 16 of 16 text datasets, Solar Decide Flash on 0. Same cases, same request, same scorer.
| Microsoft-Decision-1 | Solar Decide Flash | |
|---|---|---|
| System One Index | 82.0 | 63.6 |
| English | 81.0 | 61.3 |
| Arabic | 83.2 | 66.5 |
| Vision Index | – | – |
| Calibration error | 2.8% | 14.4% |
| Median latency | 240 ms | 586 ms |
| Cost per 1,000 calls | $0.010 | $0.042 |
| Microsoft-Decision-1 | Solar Decide Flash | |
|---|---|---|
| AG News English | 89.0 | 86.0 |
| SST-2 English | 95.5 | 90.5 |
| SST-5 English | 62.5 | 44.5 |
| DAIR Emotion English | 57.0 | 54.5 |
| Banking77 English | 87.0 | 1.0 |
| BoolQ English | 88.5 | 86.0 |
| XNLI English English | 89.3 | 65.3 |
| Typed decisions English | 79.0 | 62.5 |
| XNLI French French | 82.0 | 61.3 |
| XNLI Arabic Arabic | 72.0 | 52.0 |
| MASSIVE scenario Arabic | 83.5 | 46.0 |
| MASSIVE intent Arabic | 89.0 | 63.0 |
| OSACT4 offensive Arabic | 88.5 | 84.5 |
| Tweet sentiment Arabic | 68.5 | 65.5 |
| MIRACL rerank Arabic | 90.0 | 71.5 |
| MIRACL pairwise Arabic | 91.0 | 83.0 |
Model pages: Microsoft-Decision-1 · Solar Decide Flash. Latency for API models is the median per call through OpenRouter, timed from one machine.