GPT-3.5 Turbo (Artificial Analysis source-unspecified snapshot)
OpenAI · 2022-11-30
Artificial Analysis reports this result under the generic GPT-3.5 Turbo label and the original ChatGPT launch date without identifying a dated API checkpoint. It is kept separate from the 0301, 0613, 1106, and 0125 API snapshots.
Benchmark scores
| Benchmark | Score |
|---|---|
| Artificial Analysis Coding Index | 10.7 |
| Artificial Analysis Intelligence Index | 3.6 |
| GPQA Diamond (Artificial Analysis avg@5) | 29.7 |
| MATH-500 (Artificial Analysis source) | 44.1 |
| MMLU Pro | 46.2 |