Superficial Logo

Superficial

Gemini 2.5 Flash Lite Benchmarks

Benchmarks·June 18th 2025·2 min read
Gemini 2.5 Flash Lite Benchmarks

Google Gemini 2.5 Flash Lite Superfacts Benchmark results are in.

Highlights:

  • Gemini 2.5 Flash Lite comes in first in our accuracy benchmark with a Superfacts score of 93.35% and a FACTS score of 96.64%.
  • All three Gemini 2.5 models (Flash Lite, Flash, Pro) are now top of the Superfacts benchmark above every model from Anthropic, xAI, Deepseek and OpenAI.
  • Gemini 2.5 Flash Lite joins Gemini 2.5 Flash in achieving a perfect 100% FACTS score after one-shot enhancement with Superficial.

About Superfacts

Superfacts is the first claim-level hallucination benchmark for top AI models. Visit the benchmark at www.benchmarks.superficial.org.

About Gemini 2.5 Flash Lite

Gemini 2.5 Flash Lite is a thinking model optimized for cost-efficiency and high throughput. Learn more at deepmind.google/models/gemini/flash-lite/.