Skip to main content
Back to Explore

Gemini 2.5 Flash Preview (Sep '25) (Reasoning)

Google·Released 2025-09-25
MoEMultimodal
Compare

Related Models

Google: Gemini 3.5 Flash2026-05-19Gemini 3.5 Flash (medium)2026-05-19Gemini 3.5 Flash (minimal)2026-05-19Google: Gemini 3.1 Flash Lite2026-05-07Google: Gemini 3.1 Flash Lite Preview2026-03-03Google: Gemini 3.1 Pro Preview Custom Tools2026-02-25Google: Gemini 3.1 Pro Preview2026-02-19Gemini 3 Deep Think2026-02-05

Benchmarks

MMLU-Pro
84.2%
GPQA Diamond
79.3%
HLE
12.7%
LiveCodeBench
71.3%
SciCode
40.5%
TerminalBench Hard
16.7%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025
78.3%
IFBench
52.3%
Long Context Recall
64.3%
Tau2
45.6%
Market AverageTop Score

Quick Compare

Similar Models

OpenAI: gpt-oss-120b

OpenAI

Q: 23.8$0.06/1M131K ctx
Coding: +5.8

Claude 4.5 Haiku (Non-reasoning)

Anthropic

Q: 23.7$2.00/1M
Coding: +5.0

Qwen: Qwen3 Max

Alibaba

Q: 24.0$1.56/1M262K ctx

MoonshotAI: Kimi K2 0905

Kimi

Q: 23.5$1.07/1M262K ctx

Claude 3.7 Sonnet (Non-reasoning)

Anthropic

Q: 23.5$6.00/1M200K ctx

Qwen3.6 35B A3B (Non-reasoning)

Alibaba

Q: 24.2$0.84/1M
Coding: -7.0
Compare all 7 models