| Spec | Gemini 2.5 Pro | o3 |
|---|
| Provider | OpenAI | |
| Release Date | May 2025 | Apr 2025 |
| Knowledge Cutoff | 2025-01-31 | 2024-05-31 |
| Parameters | Undisclosed | Undisclosed |
| Context Window | 1.0M | — |
| Max Output | — | — |
| Open Source | No | No |
| License | proprietary | proprietary |
| Tokenizer | — | — |
| Modality | — | — |
| Reasoning Model | No | No |
| Moderated | No | No |
| Price (per 1M tokens) | Gemini 2.5 Pro | o3 |
|---|---|---|
| Input | $1.25 | — |
| Output | $10.00 | — |
| Feature | Gemini 2.5 Pro | o3 |
|---|---|---|
| Text Input | ||
| Image Input (Vision) | ||
| Audio Input | ||
| Video Input | ||
| File Input | ||
| Image Output | ||
| Audio Output | ||
| Tool Use | ||
| Structured Output (JSON) | ||
| Streaming | ||
| Reasoning Tokens | ||
| Web Search | ||
| Temperature Control | ||
| Top-P Sampling | ||
| Stop Sequences | ||
| Seed (Reproducibility) | ||
| Log Probabilities |
| Benchmark | Gemini 2.5 Pro | o3 |
|---|---|---|
| SimpleQA | 50.8 | — |
| Benchmark | Gemini 2.5 Pro | o3 |
|---|---|---|
| SWE-bench Verified | 63.2 | 69.1 |
| Benchmark | Gemini 2.5 Pro | o3 |
|---|---|---|
| AIME 2025 | 83 | 86.4 |
| FrontierMath | — | 15.8 |
| Benchmark | Gemini 2.5 Pro | o3 |
|---|---|---|
| GPQA Diamond | 83 | 83.3 |
| Humanity's Last Exam | 17.8 | 14.7 |
| ARC-AGI v2 | 4.9 | 6.5 |
| Benchmark | Gemini 2.5 Pro | o3 |
|---|---|---|
| MMMU | 79.6 | 82.9 |
| MMMU-Pro | — | 76.4 |
| CharXiv Reasoning | — | 78.6 |
| Benchmark | Gemini 2.5 Pro | o3 |
|---|---|---|
| BrowseComp | — | 49.7 |