Provider logo

Gemini 2.5 Flash Lite Preview (09/2025) – Thinking

gemini-2.5-flash-lite-preview-09-2025-thinking
Provider logo

Gemini 2.5 Flash Lite Preview (09/2025) – Thinking

gemini-2.5-flash-lite-preview-09-2025-thinking

Deprecated compatibility alias. Requests route to Gemini 2.5 Flash Lite with the stable thinking path.

Added Sep 25, 2025

Context Window

1.0M

Max Output

65.5K

Input Price (Auto)

$0.10/1M

Output Price (Auto)

$0.40/1M

Cache Read (Auto)

$0.010/1M

Capabilities

Benchmarks

Performance metrics and benchmarks

Sourced from Artificial Analysis.

Intelligence Index

15.2

Better than 45% of models compared

Reasoning

GPQA Diamond

Graduate-level scientific reasoning

70.9%

Better than 52% of models compared

HLE

Humanity's Last Exam

7.0%

Better than 48% of models compared

IFBench

Instruction-following benchmark

52.6%

Better than 63% of models compared

T²-Bench Telecom

Conversational AI agents in dual-control scenarios

30.7%

Better than 38% of models compared

AA-LCR

Long context reasoning evaluation

63.0%

Better than 67% of models compared

CritPt

Research-level physics reasoning

0.0%

Coding

SciCode

Python programming for scientific computing

28.7%

Better than 38% of models compared

Terminal-Bench Hard

Agentic coding and terminal use

12.9%

Better than 49% of models compared

LiveCodeBench

Contamination-free coding benchmark

68.8%

Better than 77% of models compared

Math

AIME 2025

American Invitational Mathematics Examination 2025

68.7%

Better than 63% of models compared

Knowledge

MMLU-Pro

Professional and academic subject knowledge

80.8%

Better than 71% of models compared

AA-Omniscience Accuracy

Proportion of correctly answered questions

18.0%

AA-Omniscience Hallucination Rate

Rate of incorrect answers among non-correct responses

87.8%

Last updated Aug 16, 2026

Artificial Analysis

Providers

Auto routing is available for this model. Explicit provider selection is not available.

Loading provider options…

Compare Gemini 2.5 Flash Lite Preview (09/2025) – Thinking with similar models from the same provider or model family.