GPT-4o
OpenAI·
LLMsflagshipcloud
Overview
Omni model accepting text, audio, and image inputs.
Capabilities and innovations
Real-time voiceNative multimodalFaster inferenceNative Omni-modalReal-time Voice (320ms)End-to-end Multimodal Training
Benchmarks
| Benchmark | Score | Source |
|---|---|---|
| GPQA Diamond | 53.6% | unsourced |
| Humanity’s Last Exam (no tools) | 8.5% | unsourced |
| SWE-bench Verified | 38% | unsourced |
| MMLU | 88.7% | 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.