Gemini 2.5 Flash (Reasoning)
GoogleGemini
描述
A thinking model designed for a balance between price and performance. It builds upon Gemini 2.0 Flash with upgraded reasoning, hybrid thinking control, multimodal capabilities (text, image, video, audio input), and a 1M token input context window.
发布日期
2025-05-20
参数规模
—
上下文长度
1.0M
支持模态
audio, image, pdf, text, video
能力雷达图
38
general
63
coding
80
reasoning
52
science估算
75
agents
80
multimodal
Science 在缺少专门科学评测时使用推理能力代理估算。
排行榜排名
基准测试分数 (LLM Stats)
Biology
GPQA
82.8%自报
Code
Aider-Polyglot
61.9%自报
SWE-Bench Verified
60.4%自报
Aider-Polyglot Edit
56.7%自报
Factuality
FACTS Grounding
85.3%自报
SimpleQA
26.9%自报
General
Global-MMLU-Lite
88.4%自报
MMMU
79.7%自报
Vibe-Eval
65.4%自报
LiveCodeBench v5
63.9%自报
MRCR
32.0%自报
Math
AIME 2024
88.0%自报
AIME 2025
72.0%自报
Humanity's Last Exam
11.0%自报
AA 评测指数
Math Index73.3
Intelligence Index20.1
Math 5001.0
Mmlu Pro0.8
Aime0.8
Gpqa0.8
Aime 250.7
Livecodebench0.7
Lcr0.6
Ifbench0.5
Scicode0.4
Tau20.3
Terminalbench Hard0.1
Hle0.1
LLM Stats 分类评分
Language90
Grounding90
Physics80
Healthcare80
Biology80
Chemistry80
Multimodal70
Math60
Reasoning60
Factuality60
Frontend Development60
General60
Code60
Vision50
Long Context20
定价
输入价格$0.3 / 1M tokens
输出价格$2.5 / 1M tokens
混合价格(3:1)$0.85 / 1M tokens
缓存读取价格$0.03 / 1M tokens
速度
Tokens/秒0.0
首Token延迟0.00s
首回答延迟0.00s
供应商价格排行
供应商价格排行
1 个供应商
供应商输入输出
1Google主要
$0.3
$2.5
比较该模型在不同 API 供应商之间的定价。