OpenAI
GPT-3.5 Turbo (Nov 2023) benchmark results
6 results on 5 benchmarks from 3 sources.
Closed Released: Nov 6, 2023 Updated Oct 22, 2026
Also listed as: gpt-3.5-turbo-1106 · GPT-3.5 version 1106 · GPT-3.5-Turbo (Nov 2023)
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1204 | 343 | Overall, style control on; 16619 votes; 95% CI 1195–1213“gpt-3.5-turbo-1106” | Official leaderboard | Oct 8, 2026 |
| GPQA Diamond | 28.0% | 265 | Epoch AI run, accuracy; effort: Nov 2023 | Independent run | Jan 27, 2025 |
| MMLU | 71.4% | 22 | Exact match; 5-shot; effort: Nov 2023 | Paper / tech report via Epoch AI | Nov 6, 2023 |
| HumanEval / HumanEval+ | 76.8% | 25 | HumanEval (base tests), pass@1 greedy“GPT-3.5-Turbo (Nov 2023)” | Official leaderboard | Dec 26, 2024 |
| HumanEval / HumanEval+ | 70.7% | 30 | HumanEval+ (extended tests), pass@1 greedy“GPT-3.5-Turbo (Nov 2023)” | Official leaderboard | Dec 26, 2024 |
| MATH Level 5 | 15.9% | 90 | Epoch AI run, accuracy; effort: Nov 2023 | Independent run | Jan 27, 2025 |
FAQ
Where do these GPT-3.5 Turbo (Nov 2023) scores come from?
From Epoch AI Benchmarking Hub, EvalPlus leaderboard, and LMArena leaderboard dataset. Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.