GPT-5.5 (Medium): benchmark results

GPT-5.5 evaluated at the medium reasoning-effort setting. Provider: OpenAI. Released 2026-04-23. Access: API.

Unified ELO 1718 ± 1, rank #67 of 1761 rated models, from 46 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
UGI - Natural Intelligence77.12NatInt Score99.8
LLM Chess (Saplin)1532.2ELO98.8
AA Omniscience - Software Engineering (SWE)85.5Accuracy (%)97.4
AA Terminal-Bench Hard57.58Accuracy (%)97.3
AA Long Context Reasoning83Accuracy (%)97
HalluHard0.42Turn-1 Hallucination Rate96.8
LisanBench0.64Mean Path Length / Current Maximum96.1
AA Omniscience - Health49.75Accuracy (%)95.9
AA Omniscience - Business47.6Accuracy (%)95.7
AA Omniscience - Humanities & Social Sciences54.3Accuracy (%)95.7
AA-Omniscience Accuracy56.8Accuracy (%)95.7
UGI - Writing64.75Writing Score95.4

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