跳转到主要内容

Gemini 3 Flash Preview (Non-reasoning)

GoogleGeminiProprietary

描述

Gemini 3 Flash offers frontier intelligence built for speed at a fraction of the cost. It combines Gemini 3's Pro-grade reasoning with Flash-level latency, efficiency and cost. Features a 1 million-token input context window and is optimized for agentic workflows, coding, and complex analysis.

发布日期
2025-12-17
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

100
general
100
coding
80
reasoning
77
science估算
60
agents
80
multimodal

缺少专门科学评测时,Science 由 LLM Stats 科学得分或推理能力估算。

排行榜排名

领域#排名分数来源
智能体能力模型榜46
54.0
LS
多模态榜52
57.0
LS

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Agents

t2-bench90.2%自报
Toolathlon49.4%自报
Finance Agent v242.5%

Factuality

SimpleQA68.7%自报

Language

MMMLU91.8%自报

Legal

Legal Agent Benchmark0.0%

Long Context

MRCR v2 (8-needle)22.1%自报

Math

AIME 202599.7%自报
LiveBench72.4%

Multimodal

VideoMMMU86.9%自报

Reasoning

Vending-Bench 2363500.0%自报
LiveCodeBench Pro2316.00 / 3000自报
Global PIQA92.8%自报
GPQANYU + Cohere + Anthropic (2023)90.4%自报
CharXiv-R80.3%自报
SWE-Bench Verified78.0%自报
FACTS Grounding61.9%自报
MCP Atlas57.4%自报
Terminal-Bench 2.0Stanford × Laude Institute (2026)47.6%自报
Humanity's Last Exam43.5%自报
ARC-AGI v233.6%自报

Vision

MMMU-Pro81.2%自报
ScreenSpot Pro69.1%自报
OmniDocBench 1.512.1%自报

AA 评测指数

(Artificial Analysis)

暂无 AA 评测数据

LLM Stats 分类评分

(LLM Stats (zeroeval))
Code
100
Agents
100
Reasoning
100
General
100
Language
90
Physics
90
Biology
90
Chemistry
90
Math
80
Frontend Development
80
Multimodal
70
Factuality
70
Grounding
70
Tool Calling
60
Vision
60
Spatial Reasoning
50
Healthcare
50
Finance
40
Long Context
20
Structured Output
10
Legal
0

定价

输入价格$0.5 / 1M tokens
输出价格$3 / 1M tokens
混合价格(3:1)$1.125 / 1M tokens
缓存读取价格$0.05 / 1M tokens

速度

Tokens/秒196.3
首Token延迟0.77s
首回答延迟0.77s

供应商价格排行

供应商价格排行

3 个供应商

最便宜: Google最贵: Neon
供应商输入输出
1Google主要
$0.5
$3
2OpenCode
$0.5
$3
3Neon
$0.5
$3

比较该模型在不同 API 供应商之间的定价。

外部链接