Dots3-Note (Preview): benchmark results
Provider: dots studio. Released 2026-08-14. Access: Open.
Aggregate Rating 1731 ± 34, rank #231 of 10916 rated models, from 12 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| The Aggregate - Inversion and Execution Ladder | 100 | Accuracy (%) | 97.6 |
| ARC-AGI-2 | 76.81 | Accuracy (%) | 83.7 |
| BenchLM | 64.1 | Overall Score | 82.8 |
| Humanity's Last Exam (Self-Reported, With Tools) | 52.6 | Accuracy (%) | 66.7 |
| WideResearch (BenchLM) | 78.9 | Task Score (%) | 60 |
| Tinybird AI SQL Benchmark - Success Rate | 98 | Questions answered with a valid query within 3 retries (%) | 35.1 |
| LLM2014 Logic 2026-08 | 12.95 | Median Score | 20.5 |
| Tinybird AI SQL Benchmark - Exactness | 39.27 | Result exactness vs human reference queries (0-100) | 17 |
| LLM2014 Logic 2026-09 | 10.32 | Median Score | 16.4 |
| The Aggregate - Day-one Recall Check | 11.8 | Accuracy (%) | 16.2 |
| LLM2014 Logic 2026-10 | 4.91 | Median Score | 10.5 |
| Tinybird AI SQL Benchmark - First-Attempt Success Rate | 66 | Questions answered with a valid query on the first attempt ( | 10.3 |
Interactive version: theaggregate.ai/model?slug=dots3-note-preview · How It Works · Data refreshed daily, snapshot 2026-11-08.