All comparisons
HumanEval 93.1 96.7 SWE-bench Verified 63.2 71.7 LiveCodeBench 70.4 79.1 Chatbot Arena 1398.0 1412.0 Google-Proof Q&A 86.4 87.7 MMLU PRO 86.2 — Aider Polyglot — 81.3 Massive Multi-discipline Multimodal Understanding — 82.9
Gemini 2.5 Pro vs o3
Side-by-side comparison of specifications, capabilities, and pricing.
| Released | 2025-06-17 | 2025-03-31 |
| Context window | 1M | 128K |
| Parameters | — | — |
| Architecture | decoder only | decoder only |
| License | Proprietary | Unknown |
| Knowledge cutoff | 2025-01 | — |
Capabilities | ||
| Vision | ||
| Multimodal | ||
| Reasoning | ||
| Function calling | ||
| Tool use | ||
| Structured Outputs | ||
| Code execution | ||
Availability | ||
| Providers | ||
Benchmarks