GPT-5 (Low) — benchmark results

GPT-5 evaluated at the low reasoning-effort setting. Provider: OpenAI. Released 2025-08-07. Access: API.

Unified ELO 1807 ± 23, rank #136 of 1776 rated models, from 63 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
UGI - Natural Intelligence67.2NatInt Score97.3
AA MATH-50098.73Accuracy (%)96
AA MMLU-Pro85.98Accuracy (%)93.8
GAIA2 - Ambiguity39.6Score (%)93.3
GAIA2 - Search64.2Score (%)93.3
Aider polyglot coding leaderboard81.3Pass rate (%)93.2
LLM Chess (Saplin)858.2ELO92.1
AA Omniscience - Health38.6Accuracy (%)91.7
UGI - Writing48.31Writing Score88.2
AA Omniscience - Law32.1Accuracy (%)87.5
AA LiveCodeBench76.3Pass@1 (%)87
AA Omniscience - Software Engineering (SWE) - Julia40Accuracy (%)86.8

Interactive version: theaggregate.ai/model?slug=gpt-5-low · How the rankings work · Data refreshed daily, snapshot 2026-07-22.