Gemini 2.5 Flash Lite
von Google DeepMind · Veroeffentlicht 2025-07-22
59.1
Durchschn. Score
$0.10/1M
Eingabepreis
$0.40/1M
Ausgabepreis
1.0M tokens (~524 books)
Kontextfenster
multimodal
Typ
Tested on 8 benchmarks with 59.1% average. Top scores: HELM — WildBench (81.8%), HELM — IFEval (81.0%), HELM — MMLU-Pro (53.7%).
Benchmark-Ergebnisse
| Benchmark | Kategorie | Score | Bar |
|---|---|---|---|
| HELM — WildBench | reasoning | 81.8 | |
| HELM — IFEval | language | 81.0 | |
| HELM — MMLU-Pro | knowledge | 53.7 | |
| HELM — Omni-MATH | math | 48.0 | |
| HELM — GPQA | knowledge | 30.9 | |
| Artificial Analysis — Quality Index | speed | 21.6 | |
| Artificial Analysis — Coding Index | speed | 18.1 | |
| Artificial Analysis — Agentic Index | speed | 11.7 |