TPSTokens Per Second

Gemini 3.1 Flash Lite Preview

Google DeepMind🇺🇸 United States
active

Google's high-efficiency model optimized for high-volume use cases. Outperforms Gemini 2.5 Flash Lite on overall quality and approaches Gemini 2.5 Flash performance. Improvements span audio input/ASR and RAG.

Compare with other models →
Context window1050K tokens
Input / 1M tokens$0.25
Output / 1M tokens$1.5

Version History

gemini-3.1-flash-lite-preview-2026-03-02major

Gemini 3.1 Flash Lite Preview launches as Google's high-efficiency model for high-volume use. 1M context at $0.25/$1.50, outperforms Gemini 2.5 Flash Lite.

Benchmark Scores

Full leaderboard →
1432.0 elo
Arena Elo
82.2%
GPQA
8.2%
Hallucination (Summarization)
319.0 tokens_per_sec
Speed (tok/s)