Gemini 2.5 Flash
por Google DeepMind · Lançado em 2025-06-17
43.3
pontuação média
$0.30/1M
Preço de entrada
$2.50/1M
Preço de saída
1.0M tokens (~524 books)
Janela de contexto
multimodal
Tipo
Tested on 10 benchmarks with 43.3% average. Top scores: Lech Mazur Writing (76.5%), OTIS Mock AIME 2024-2025 (73.0%), GeoBench (73.0%).
Pontuações de benchmark
| Benchmark | Categoria | Pontuação | Bar |
|---|---|---|---|
| Lech Mazur Writing | knowledge | 76.5 | |
| OTIS Mock AIME 2024-2025 | math | 73.0 | |
| GeoBench | knowledge | 73.0 | |
| Fiction.LiveBench | knowledge | 47.2 | |
| Aider polyglot | coding | 47.1 | |
| WeirdML | coding | 41.0 | |
| ARC-AGI | reasoning | 32.3 | |
| SWE-Bench Verified (Bash Only) | coding | 28.7 | |
| HLE | knowledge | 7.7 | |
| VPCT | knowledge | 7.0 |