Dolphin-1.5: benchmark results
Provider: Cognitive Computations.
Aggregate Rating 1633 ± 51, rank #691 of 10552 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SEA-Vision - Document Parsing (Vietnamese) | 0.17 | Normalized edit distance (0-1) between the parsed page and t | 58.3 |
| SEA-Vision - Document Parsing (Burmese) | 0.48 | Normalized edit distance (0-1) between the parsed page and t | 54.2 |
| SEA-Vision - Document Parsing (Portuguese) | 0.15 | Normalized edit distance (0-1) between the parsed page and t | 50 |
| SEA-Vision - Document Parsing | 0.32 | Normalized edit distance (0-1) between the parsed page and t | 41.7 |
| SEA-Vision - Document Parsing (English) | 0.16 | Normalized edit distance (0-1) between the parsed page and t | 41.7 |
| SEA-Vision - Document Parsing (Filipino) | 0.09 | Normalized edit distance (0-1) between the parsed page and t | 41.7 |
| SEA-Vision - Document Parsing (Indonesian) | 0.17 | Normalized edit distance (0-1) between the parsed page and t | 41.7 |
| SEA-Vision - Document Parsing (Khmer) | 0.69 | Normalized edit distance (0-1) between the parsed page and t | 41.7 |
| SEA-Vision - Document Parsing (Thai) | 0.55 | Normalized edit distance (0-1) between the parsed page and t | 41.7 |
| OmniDocBench v1.6 Full - Reading Order Edit | 0.17 | Normalized edit distance | 31.4 |
| OmniDocBench 1.5 | 85.07 | Overall (mixed sources) | 27.9 |
| SEA-Vision - Document Parsing (Chinese) | 0.2 | Normalized edit distance (0-1) between the parsed page and t | 25 |
Interactive version: theaggregate.ai/model?slug=dolphin-1-5 · How It Works · Data refreshed daily, snapshot 2026-11-16.