Gemini 2.5 Deep Think model card 1 Aug 2025
Gemini 2.5 Deep Think
Gemini 2.5 Deep Think places in the top three on MMMU.
8
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Google DeepMind Aug 2025 Gemini 2.5 Deep Think
Closed weights Flagship
No Epoch Capabilities Index score yet
- IMO 2025 60.7% #6 /6
- FrontierMath 29% #19 /70
- FrontierMath Tier 4 10.4% #20 /55
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Humanity's Last Exam | 34.8% | #67 of 147 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIME 2025 | 99.2% | #11 of 143 | Model card | |
| FrontierMath | 29% | #19 of 70 | Epoch AI | |
| FrontierMath Tier 4 | 10.4% | #20 of 55 | Epoch AI | |
| IMO 2025 | 60.7% | #6 of 6 | Model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v6 | 87.6% | #13 of 76 | Model card |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMMU | 84% | #3 of 136 | Official leaderboard | |
| VoxelBench | 1,472 | #39 of 56 | Official leaderboard |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.