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 1434 ± 1, rank #1015 of 1392 rated models, from 33 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ClassEval | 29.6 | Class-Level Pass@1 (%) | 90 |
| CodeScope | 35.14 | Score | 85.7 |
| CodeScope - Code Generation | 21.07 | Score | 85.7 |
| CodeScope - Code Repair | 13.54 | Score | 85.7 |
| CodeScope - Code Summarization | 33.18 | Score | 85.7 |
| CodeScope - Code Translation | 21.37 | Score | 85.7 |
| CodeScope - Code Understanding | 49.2 | Score | 85.7 |
| CodeScope - Program Synthesis | 22.91 | Score | 85.7 |
| InterCode | 34.5 | Success Rate (%) | 75 |
| PubMedQA | 79.6 | Accuracy (%) | 72.7 |
| EvalPlus (HumanEval+ & MBPP+) | 66.5 | Pass@1 avg (%) | 69.8 |
| Nexus Function Calling - VirusTotal | 81 | Exact Match (%) | 66.7 |
Interactive version: theaggregate.ai/model?slug=gpt-3-5 · How It Works · Data refreshed daily, snapshot 2026-09-05.