GPT-4
OpenAI·
LLMsflagshipcloud
Overview
Multimodal model exhibiting human-level performance on various professional and academic benchmarks.
Capabilities and innovations
Multimodal (Image inputs)Advanced reasoning32k context windowNative Multimodal InputSystem Card SafetyAdvanced Vision Understanding
Benchmarks
| Benchmark | Score | Source |
|---|---|---|
| GPQA Diamond | 35.7% | unsourced |
| Humanity’s Last Exam (no tools) | 4.2% | unsourced |
| SWE-bench Verified | 1.7% | unsourced |
| MMLU | 86.4% | unsourced |
- SWE-bench Pro: benchmark did not exist at release
- MMLU-Pro: benchmark did not exist at release
- Terminal-Bench 2.x: benchmark did not exist at release
- MMMU-Pro: benchmark did not exist at release
Links
More from OpenAI
Data curated by AI Model Timeline. See the methodology for admission criteria, benchmark eras and source priority.