Gemini 2.5 Flash Lite Preview 09-25

Google DeepMind · google/gemini-2-5-flash-lite-preview-09-2025

ProviderGoogle DeepMind
Familygemini-flash-lite
Typellm-reasoning
Statuspreview
Released2025-09-25
Updated2025-09-25

Reported benchmark scores

These scores carry one collection date for the whole card (2026-04) and a source list rather than a source per score, so they cannot be attributed individually. They are shown as reported and are not verified evidence. Benchmark pages carry the reviewed, dated evidence where it exists.
BenchmarkCatalogue standingScore
Aider Polyglot Benchmarkunassessed55.1
AIME 2025unassessed73.3
Arena Elo — Codingunassessed1330.0
Arena Elo — Hard Promptsunassessed1390.7
Arena Elo — Mathunassessed1350.0
Arena Elo — Overall (Text)unassessed1340.0
Arena Elo — Style Controlunassessed1380.0
Arena Elo — Visionunassessed1173.2
Artificial Analysis Intelligence Indexunassessed80.0
Artificial Analysis Output Speedunassessed87.0
GPQA Diamondunassessed70.2
HumanEvalunassessed88.8
IFEvalunassessed87.5
LiveCodeBenchunassessed33.7
MATH-500unassessed90.5
MMLU-Prounassessed78.2
MultiPL-E: C#unassessed78.5
MultiPL-E: Juliaunassessed60.5
MultiPL-E: Kotlin (unconfirmed)unassessed71.5
MultiPL-E: Luaunassessed53.8
MultiPL-E: Perlunassessed49.5
MultiPL-E: PHPunassessed76.2
MultiPL-E: Runassessed56.2
MultiPL-E: Rubyunassessed64.5
MultiPL-E: Scalaunassessed58.8
MultiPL-E: Swiftunassessed64.8
SWE-bench Verifiedunassessed49.2

Data

This card as JSON · Edit on GitHub