GPT-5.6 Sol (Non-reasoning) — benchmark results

Provider: OpenAI. Released 2026-07-09. Access: API.

Unified ELO 1867 ± 31, rank #88 of 1744 rated models, from 36 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA Omniscience - Software Engineering (SWE) - Swift84Accuracy (%)98.1
UGI - Natural Intelligence70.1NatInt Score98
AA Omniscience - Software Engineering (SWE) - Go80Accuracy (%)97.9
AA Omniscience - Software Engineering (SWE) - Kotlin70Accuracy (%)96.4
AA Omniscience - Software Engineering (SWE) - TypeScript78.89Accuracy (%)96.2
AA Omniscience - Health44.7Accuracy (%)96.1
AA Omniscience - Software Engineering (SWE) - Rust80Accuracy (%)96.1
AA Omniscience - Software Engineering (SWE)72.1Accuracy (%)96
AA Omniscience - Software Engineering (SWE) - Python78Accuracy (%)96
AA Omniscience - Business41.7Accuracy (%)95.9
AA-Omniscience Accuracy48.18Accuracy (%)95.7
AA Omniscience - Law42.1Accuracy (%)95.2

Interactive version: aibenchmarks.dev/model?slug=gpt-5-6-sol-non-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-10.