GPT-5.6 Sol (Low): benchmark results

GPT-5.6 Sol evaluated at the low reasoning-effort setting. Provider: OpenAI. Released 2026-07-09. Access: API.

Unified ELO 1701 ± 1, rank #104 of 1761 rated models, from 31 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA Terminal-Bench Hard60.61Accuracy (%)98.5
AA Omniscience - Health51.6Accuracy (%)97.1
AA-Omniscience Accuracy57.17Accuracy (%)96.3
AA Omniscience - Business48Accuracy (%)95.9
AA Omniscience - Science, Engineering & Mathematics53.2Accuracy (%)95.9
AA Omniscience - Software Engineering (SWE)83Accuracy (%)95.5
AA Omniscience - Humanities & Social Sciences54.1Accuracy (%)95
AA Omniscience - Law53.1Accuracy (%)94
AA Omniscience18.87Score91.4
Artificial Analysis Intelligence Index40.84Intelligence Index90.5
AA Humanity's Last Exam39.39Accuracy (%)90.3
AA GPQA Diamond89.8Accuracy (%)88.9

Interactive version: theaggregate.ai/model?slug=gpt-5-6-sol-low · How It Works · Data refreshed daily, snapshot 2026-09-05.