跳轉到主要內容

Gemini 3 Pro Preview (low)

GoogleGeminiProprietary

描述

Gemini 3 Pro is the first model in the new Gemini 3 series. It is best for complex tasks that require broad world knowledge and advanced reasoning across modalities. Gemini 3 Pro uses dynamic thinking by default to reason through prompts, and features a 1 million-token input context window with 64k output tokens.

發布日期
2025-11-18
參數規模
上下文長度
支援模態
audio, image, text, video

能力雷達圖

54
general
57
coding
87
reasoning
63
science估算
70
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

領域#排名分數來源
智能体与工具6
75.0
LS
代码能力榜42
74.0
AA
通用能力榜82
71.0
AA
数学推理51
88.0
AA
多模态榜70
42.0
LS
推理能力99
31.0
LS
科学能力24
80.0
AA

基準測試分數 (LLM Stats)

Agents

Vending-Bench 2547816.0%自報
t2-bench85.4%自報
Terminal-Bench 2.054.2%自報

Biology

GPQA91.9%自報

Code

LiveCodeBench Pro2439.00 / 3000自報
SWE-Bench Verified76.2%自報

Factuality

SimpleQA72.1%自報
FACTS Grounding70.5%自報

General

Global PIQA93.4%自報
MMMLU91.8%自報
MMMU-Pro81.0%自報
MRCR v2 (8-needle)26.3%自報

Grounding

ScreenSpot Pro72.7%自報

Healthcare

VideoMMMU87.6%自報

Math

AIME 2025100.0%自報
Humanity's Last Exam45.8%自報
MathArena Apex23.4%自報

Multimodal

CharXiv-R81.4%自報
OmniDocBench 1.511.5%自報

Reasoning

ARC-AGI v231.1%自報

AA 評測指數

Math Index
86.7
Intelligence Index
41.3
Coding Index
39.4
Mmlu Pro
0.9
Gpqa
0.9
Aime 25
0.9
Livecodebench
0.9
Tau2
0.7
Lcr
0.7
Scicode
0.5
Ifbench
0.5
Terminalbench Hard
0.3
Hle
0.3

LLM Stats 分類評分

Agents
100
Code
100
Reasoning
100
General
100
Biology
90
Chemistry
90
Healthcare
90
Language
90
Physics
90
Frontend Development
80
Tool Calling
70
Factuality
70
Grounding
70
Math
70
Multimodal
70
Vision
60
Spatial Reasoning
50
Long Context
30
Structured Output
10

定價

輸入價格$2 / 1M tokens
輸出價格$12 / 1M tokens
混合價格(3:1)$4.5 / 1M tokens

速度

Tokens/秒0.0 tokens/s
首Token延遲0.00s
首回答延遲0.00s

可用提供商

(LS 內部計價單位)

暫無提供商資料

外部連結