Trinity-Large-Thinking: benchmark results
Arcee Trinity Large evaluated with thinking enabled. Provider: Arcee AI. Released 2026-04-01. Access: Open.
Aggregate Rating 1630 ± 15, rank #356 of 1656 rated models, from 154 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA TAU-2 Bench | 90.06 | Accuracy (%) | 84.6 |
| Tau2-Bench Telecom | 90.06 | Success Rate (%) | 84.4 |
| AA Omniscience - Software Engineering (SWE) - R | 16 | Accuracy (%) | 72.5 |
| Tinybird AI SQL Benchmark - Success Rate | 100 | Questions answered with a valid query within 3 retries (%) | 71.7 |
| AA Omniscience - Software Engineering (SWE) - Java | 19 | Accuracy (%) | 71 |
| AA Omniscience - Software Engineering (SWE) - Swift | 40 | Accuracy (%) | 70.1 |
| Arcee AIME 2025 (BenchLM) | 96.3 | Accuracy (%) | 70 |
| AA IFBench | 56.26 | Accuracy (%) | 67.3 |
| AA Omniscience - Software Engineering (SWE) - PHP | 26 | Accuracy (%) | 65 |
| AA Omniscience - Software Engineering (SWE) - Rust | 52 | Accuracy (%) | 64.7 |
| Position Bias (Lechmazur) | 36.6 | Order Flip % (lower is better) | 62.9 |
| AA Terminal-Bench Hard | 22.73 | Accuracy (%) | 62.6 |
Interactive version: theaggregate.ai/model?slug=trinity-large-thinking · How It Works · Data refreshed daily, snapshot 2026-10-20.