Gemini 2.5 Flash Preview (Sep '25) (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.
發布日期
2025-09-25
參數規模
—
上下文長度
1.0M
支援模態
audio, image, pdf, text, video
能力雷達圖
40
general
64
coding
78
reasoning
52
science估算
74
agents
80
multimodal
Science 在缺少專門科學評測時使用推理能力代理估算。
排行榜排名
基準測試分數 (LLM Stats)
Biology
GPQA
82.8%自報
Code
Aider-Polyglot
61.9%自報
SWE-Bench Verified
60.4%自報
Aider-Polyglot Edit
56.7%自報
Factuality
FACTS Grounding
85.3%自報
SimpleQA
26.9%自報
General
Global-MMLU-Lite
88.4%自報
MMMU
79.7%自報
Vibe-Eval
65.4%自報
LiveCodeBench v5
63.9%自報
MRCR
32.0%自報
Math
AIME 2024
88.0%自報
AIME 2025
72.0%自報
Humanity's Last Exam
11.0%自報
AA 評測指數
Math Index78.3
Intelligence Index23.8
Mmlu Pro0.8
Gpqa0.8
Aime 250.8
Livecodebench0.7
Lcr0.6
Ifbench0.5
Tau20.5
Scicode0.4
Terminalbench Hard0.2
Hle0.1
LLM Stats 分類評分
Language90
Grounding90
Physics80
Healthcare80
Biology80
Chemistry80
Multimodal70
Math60
Reasoning60
Factuality60
Frontend Development60
General60
Code60
Vision50
Long Context20
定價
輸入價格免費
輸出價格免費
混合價格(3:1)免費
快取讀取價格$0.03 / 1M tokens
速度
Tokens/秒0.0
首Token延遲0.00s
首回答延遲0.00s
供應商價格排行
暫無提供商資料