Trinity Large (Thinking) — benchmark results
Trinity Large evaluated with thinking enabled. Provider: Arcee AI. Released 2026-04-01. Access: Open.
Unified ELO 1671 ± 26, rank #319 of 1776 rated models, from 39 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA TAU-2 Bench | 90.06 | Accuracy (%) | 84.6 |
| CritPt | 0.9 | Accuracy (self-reported) | 74.5 |
| AA Humanity's Last Exam | 14.69 | Accuracy (%) | 72.7 |
| AA Omniscience - Science, Engineering & Mathematics | 33.3 | Accuracy (%) | 72.2 |
| AA CritPt | 0.86 | Accuracy (%) | 68.8 |
| AA IFBench | 56.26 | Accuracy (%) | 67.2 |
| AA Omniscience - Health | 22.4 | Accuracy (%) | 64 |
| AA Omniscience - Humanities & Social Sciences | 23.2 | Accuracy (%) | 63.1 |
| AA Terminal-Bench Hard | 22.73 | Accuracy (%) | 62.6 |
| AA GPQA Diamond | 75.15 | Accuracy (%) | 62.3 |
| AA Omniscience - Law | 14 | Accuracy (%) | 61.8 |
| AA-Omniscience Accuracy | 22.75 | Accuracy (%) | 61.4 |
Interactive version: theaggregate.ai/model?slug=trinity-large-thinking · How the rankings work · Data refreshed daily, snapshot 2026-07-22.