Agnes 2.5 Pro Alpha: benchmark results

Provider: Other. Released 2026-07-24. Access: API.

Unified ELO 1641 ± 1, rank #280 of 1761 rated models, from 18 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
BenchmarkList ECI137.02Capability Index (ECI)86.4
AA CritPt10.86Accuracy (%)86.3
AA Humanity's Last Exam33.6Accuracy (%)85.1
AA GPQA Diamond87.58Accuracy (%)84
AA Omniscience - Science, Engineering & Mathematics43.5Accuracy (%)83.7
Artificial Analysis Intelligence Index31.25Intelligence Index82.5
AA Long Context Reasoning75.67Accuracy (%)78.8
AA Omniscience - Humanities & Social Sciences33.3Accuracy (%)78.8
AA Omniscience - Health32.2Accuracy (%)77.6
AA-Omniscience Accuracy33.48Accuracy (%)77.4
AA Omniscience - Law25.1Accuracy (%)76.5
AA Omniscience - Business25.4Accuracy (%)75

Interactive version: theaggregate.ai/model?slug=agnes-2-5-pro-alpha · How It Works · Data refreshed daily, snapshot 2026-09-05.