Mercury 2.5 (High): benchmark results
Provider: Inception. Released 2026-09-08. Access: API.
Aggregate Rating 1463 ± 31, rank #1271 of 2245 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Vals AI LegalBench | 83.06 | Accuracy (%) | 59.9 |
| ObviousBench | 94.44 | Answer pass³ (%) | 55.7 |
| Vals AI CyberBench | 45.48 | Accuracy (%) | 15.6 |
| StudyArena Text - Weighted score | 1151 | Weighted score | 14.3 |
| Vals AI MedCode | 31.33 | Accuracy (%) | 11.2 |
| Vals AI Harvey Legal Agent Bench | 0 | Accuracy (%) | 10 |
| Vals AI Vibe Code Bench | 3.68 | Accuracy (%) | 8.7 |
| Vals AI Terminal-Bench 2.1 | 34.46 | Accuracy (%) | 6.7 |
| Vals AI Public Benefits Bench | 45.26 | Accuracy (%) | 6 |
| Vals AI Legal Research Bench | 4.33 | Accuracy (%) | 5.4 |
| Vals AI Code Migration | 4.52 | Accuracy (%) | 5.1 |
| Vals AI Finance Agent v2 | 18.89 | Accuracy (%) | 3.7 |
Interactive version: theaggregate.ai/model?slug=mercury-2-5-high · How It Works · Data refreshed daily, snapshot 2026-11-16.