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.

发布日期

2025-11-18

参数规模

—

上下文长度

—

支持模态

audio, image, text, video

能力雷达图

general

coding

reasoning

science

agents

multimodal

排行榜排名

领域	#排名	分数	来源
智能体能力模型榜	48	52.0	LS
代码能力榜	76	76.0	AA
通用能力榜	131	66.0	AA
多模态榜	61	41.0	LS
科学能力	69	76.0	AA

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Agents

Vending-Bench 2

547816.0%自报

t2-bench

85.4%自报

Terminal-Bench 2.0Stanford × Laude Institute (2026)

54.2%自报

Biology

GPQANYU + Cohere + Anthropic (2023)

91.9%自报

Code

LiveCodeBench Pro

2439.00 / 3000自报

SWE-Bench Verified

76.2%自报

Factuality

SimpleQA

72.1%自报

FACTS Grounding

70.5%自报

General

Global PIQA

93.4%自报

MMMLU

91.8%自报

MMMU-Pro

81.0%自报

LiveBench

73.4%

MRCR v2 (8-needle)

26.3%自报

Grounding

ScreenSpot Pro

72.7%自报

Healthcare

VideoMMMU

87.6%自报

Math

AIME 2025

100.0%自报

Humanity's Last Exam

45.8%自报

MathArena Apex

23.4%自报

Multimodal

CharXiv-R

81.4%自报

OmniDocBench 1.5

11.5%自报

Reasoning

ARC-AGI v2

31.1%自报

AA 评测指数

(Artificial Analysis)

Math Index(Artificial Analysis)

86.7

Intelligence Index(Artificial Analysis)

33.9

Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))

0.9

Gpqa(NYU + Cohere + Anthropic (2023))

0.9

Aime 25(MAA (Mathematical Association of America))

0.9

Livecodebench(UC Berkeley + MIT + Cornell (2024))

0.9

Lcr(Artificial Analysis)

0.7

Tau2(Sierra + U Toronto + Vector Institute (2025))

0.7

Scicode(UIUC + Argonne National Lab (2024))

0.5

Ifbench(Google Research (2023))

0.5

Terminalbench Hard(Stanford × Laude Institute (2026))

0.3

Hle(Center for AI Safety + Scale AI (2025))

0.3

LLM Stats 分类评分

(LLM Stats (zeroeval))

Agents

100

Code

100

Reasoning

100

General

100

Physics

Language

Healthcare

Biology

Chemistry

Frontend Development

Math

Multimodal

Factuality

Grounding

Tool Calling

Vision

Spatial Reasoning

Long Context

Structured Output

定价

输入价格$2 / 1M tokens

输出价格$12 / 1M tokens

混合价格(3:1)$4.5 / 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

1 个供应商

供应商输入输出

1Google主要

$12

比较该模型在不同 API 供应商之间的定价。

外部链接

LLM Stats Artificial Analysis