Google releases Gemini 2.5, its first "thinking" model
Gemini 2.5 Pro reasons before answering and, according to Google, leads common benchmarks by meaningful margins at launch.
Model summary
- Developer
- Google DeepMind
- Availability
- Gemini app and Google AI Studio (experimental at launch)
Reported benchmark results
| Humanity's Last Exam (no tools) | 18.8% | Highest among models without tools at the time |
Data source: Google announcement. Scores are as published by the developer at launch. Test conditions differ between companies, so compare with care.