gemini-3.1-flash-lite-preview
proprietaryMultimodal
Benchmark Average (0–100)
67.2%
Average of non-null benchmark scores across all evaluated tasks.
BENCHMARK SCORES
GPQA86.9%
Graduate-Level Google-Proof Q&A. PhD-level multiple-choice questions in chemistry, biology, and physics. Scored by accuracy.
MMMLU88.9%
Multilingual MMLU tests knowledge across many languages and subject areas.
HLE16.0%
Humanity's Last Exam (HLE) is a multi-modal benchmark testing frontier knowledge across mathematics, humanities, and natural sciences with 2,500 expert-level questions.
MMMU-Pro76.8%
MMMU-Pro is a more challenging version of MMMU with harder multimodal understanding tasks.
OTHER BENCHMARKS
SimpleQA
43.3%
CharXIV-R
73.2%
MODEL INFO
Organization
Google
Context Window
1,000,000
Release Date
2026-03-03
Knowledge Cutoff
2025-01-31
Parameters
—
License
proprietary
Input Price ($/M)
$0.2500
Output Price ($/M)
$1.5000
METADATA
Announcement Date
2026-03-03
Organization Country
US
Canonical ID
—