GPT-5.6 Sol (xHigh) — benchmark results

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

Unified ELO 2063 ± 52, rank #12 of 1744 rated models, from 39 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA Omniscience - Software Engineering (SWE) - Kotlin94Accuracy (%)100
PM-LLM-Benchmark40.2Score100
AA Omniscience - Software Engineering (SWE) - Go86Accuracy (%)99.8
Artificial Analysis Intelligence Index57.65Intelligence Index99.6
AA Omniscience - Humanities & Social Sciences57.7Accuracy (%)99.5
AA-Omniscience Accuracy57.97Accuracy (%)99.5
AA Omniscience - Software Engineering (SWE) - TypeScript91.11Accuracy (%)99.4
ARC-AGI-197.5Accuracy (%)99.4
ARC-AGI-290Accuracy (%)99.4
AA Humanity's Last Exam44.72Accuracy (%)99.3
AA Omniscience - Business49.6Accuracy (%)99.3
AA Omniscience - Health49.7Accuracy (%)99.3

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