GPT-4 Turbo (Preview): benchmark results
OpenAI preview GPT-4 Turbo model row, tracked separately from the GA GPT-4 Turbo row. Provider: OpenAI. Released 2023-11-06. Access: API.
Aggregate Rating 1614 ± 16, rank #397 of 1656 rated models, from 176 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Agent - CIBench - Chinese (CompassBench 2402) | 68.5 | Score (%) | 100 |
| OpenCompass Agent - CIBench - Chinese (CompassBench 2404) | 76.2 | Score (%) | 100 |
| OpenCompass Agent - CIBench - English (CompassBench 2402) | 82.6 | Score (%) | 100 |
| OpenCompass Agent - CIBench - English (CompassBench 2404) | 80.5 | Score (%) | 100 |
| OpenCompass Code - HumanEval - Chinese (CompassBench 2404) | 68.9 | Score (%) | 100 |
| OpenCompass Code - MBPP - English (CompassBench 2404) | 77.8 | Score (%) | 100 |
| OpenCompass LLM - Agent (CompassBench 2402) | 80.5 | Score (%) | 100 |
| OpenCompass LLM - Agent (CompassBench 2404) | 82 | Score (%) | 100 |
| OpenCompass LLM - Agent - Chinese (CompassBench 2402) | 76.7 | Score (%) | 100 |
| OpenCompass LLM - Agent - Chinese (CompassBench 2404) | 80.6 | Score (%) | 100 |
| OpenCompass LLM - Agent - English (CompassBench 2402) | 84.4 | Score (%) | 100 |
| OpenCompass LLM - Agent - English (CompassBench 2404) | 83.3 | Score (%) | 100 |
Interactive version: theaggregate.ai/model?slug=gpt-4-turbo-preview · How It Works · Data refreshed daily, snapshot 2026-10-20.