Gemini 2.5 Flash is Google's workhorse reasoning model, designed for fast, high-quality responses across coding, mathematics, and scientific tasks. It features built-in "thinking" capabilities with configurable thinking levels, allowing it to balance response speed and reasoning depth based on task complexity. Supporting a 1M-token context window with multimodal inputs including text, images, audio, video, and PDFs, it delivers strong performance at a fraction of the cost and latency of larger Gemini Pro models.
API|VisionReasoningWeb SearchFile|Proprietary Model
Knowledge Cutoff
2025-01-31
Input → Output Format
Context Memory
1.0MIN66KOUT
AI Performance Evaluation
Arena Overall Score
1411
±3As of 2026-04-23
Overall Rank
No.87
110,831 Votes
Arena by Ability
Hard Prompts
1420±3No.99
Expert Knowledge
1427±8No.87
Instruction Following
1403±4No.81
Conversation Memory
1404±5No.100
Creative
1398±5No.61
Coding
1424±5No.129
Math
1408±7No.88
Arena by Occupation
Creative Writing
1405±4No.64
Social Sciences
1430±5No.84
Media
1390±5No.75
Business
1398±5No.103
Healthcare
1426±8No.99
Legal
1428±7No.72
Software
1422±4No.117
Mathematics
1420±8No.77
Source:Arena Intelligence
Overall
AA Intelligence Index
27%↓11%
LiveBench
47%↓13%
ForecastBench
59%↓1%
Reasoning & Math
AA Math Index
73%↑0%
GPQA Diamond
79%↓2%
HLE
11%↓6%
MMLU-Pro
83%↑1%
AIME 2025
73%↑0%
MATH-500
98%↑5%
LB Reasoning
45%↓15%
LB Math
69%↓5%
LB Data
47%↓2%
Coding
AA Coding Index
22%↓12%
LiveCodeBench
70%↑4%
LB Coding
66%↓8%
LB Agentic
17%↓27%
TAU2
32%↓42%
TerminalBench
14%↓17%
SciCode
39%↓1%
Language & Instructions
IFBench
50%↓6%
AA-LCR
62%↑0%
Hallucination (HHEM)
7.8%↓2%
Factual (HHEM)
92%↑2%
LB Language
62%↓10%
LB IF
29%↓18%
Output Speed
Standard Mode
213tok/s↑131
First Output 0.44s
Reasoning Mode
211tok/s↑123
First Output 12.83s
