GPT-3.5 — benchmark results

OpenAI's ChatGPT-era GPT-3.5 chat model family (gpt-3.5-turbo lineage), the low-cost API tier that powered ChatGPT's November 2022 launch. Provider: OpenAI. Released 2022-11-30. Access: API.

Unified ELO 1426 ± 15, rank #1101 of 1776 rated models, from 33 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
TukaBench38.1ASR ↑ African Languages (self-reported)100
ClassEval29.6Class-Level Pass@1 (%)90
CodeScope35.14Score85.7
CodeScope - Code Generation21.07Score85.7
CodeScope - Code Repair13.54Score85.7
CodeScope - Code Summarization33.18Score85.7
CodeScope - Code Translation21.37Score85.7
CodeScope - Code Understanding49.2Score85.7
CodeScope - Program Synthesis22.91Score85.7
InterCode34.5Success Rate (%)75
PubMedQA79.6Accuracy (%)72.7
EvalPlus (HumanEval+ & MBPP+)66.5Pass@1 avg (%)69.8

Interactive version: theaggregate.ai/model?slug=gpt-3-5 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.