跳转到主要内容

Gemini 3.5 Flash (high)

GoogleGeminiProprietary

描述

Gemini 3.5 Flash is Google's strongest agentic and coding model in the Flash series, delivering frontier-level performance at up to 4x the speed of comparable frontier models and often at less than half the cost. Built to execute complex, agentic workflows, it outperforms Gemini 3.1 Pro on coding and agentic benchmarks while leading in multimodal understanding. It supports a 1 million-token input context window with 64k output tokens and uses dynamic thinking by default.

发布日期
2026-05-19
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

50
general
67
coding
92
reasoning
69
science
70
agents
80
multimodal

排行榜排名

领域#排名分数来源
智能体能力模型榜12
61.0
LS
代码能力榜52
85.0
AA
通用能力榜24
88.0
AA
数学推理22
88.0
LB
多模态榜21
62.0
LS
推理能力27
82.0
LB
科学能力32
87.0
AA

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Agents

Finance Agent57.9%自报
Finance Agent v257.9%
Toolathlon56.5%自报
Blueprint-Bench 233.6%自报

Code

DeepSWE 1.137.0%

Legal

Legal Agent Benchmark0.8%

Long Context

MRCR v2 (8-needle)26.6%自报

Math

LiveBench75.0%

Multimodal

OSWorld-Verified78.4%自报

Reasoning

CharXiv-R84.2%自报
MCP Atlas83.6%自报
Terminal-Bench 2.0Stanford × Laude Institute (2026)76.2%自报
ARC-AGI v272.1%自报
SWE-Bench ProPrinceton NLP (2024)55.1%自报
Humanity's Last Exam40.2%自报

Vision

MMMU-Pro83.6%自报

AA 评测指数

(Artificial Analysis)
Tau2(Sierra + U Toronto + Vector Institute (2025))
95.3
Gpqa(NYU + Cohere + Anthropic (2023))
92.2
Lcr(Artificial Analysis)
81.0
Terminalbench V2 1
78.7
Ifbench(Google Research (2023))
76.3
Coding Index(Artificial Analysis)
70.1
Scicode(UIUC + Argonne National Lab (2024))
53.1
Intelligence Index(Artificial Analysis)
52.0
Hle(Center for AI Safety + Scale AI (2025))
42.7
Terminalbench Hard(Stanford × Laude Institute (2026))
40.9
Tau Banking
32.2

LLM Stats 分类评分

(LLM Stats (zeroeval))
Multimodal
80
Spatial Reasoning
70
Tool Calling
70
Vision
70
Long Context
60
Math
60
Reasoning
60
Finance
60
General
60
Agents
60
Code
60
Legal
0

定价

输入价格$1.5 / 1M tokens
输出价格$9 / 1M tokens
混合价格(3:1)$3.375 / 1M tokens
缓存读取价格$0.15 / 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

15 个供应商

最便宜: Google最贵: Cortecs
供应商输入输出
1Google最便宜
$0
$0.00001
2UnoRouter
$0.1857
$1.1142
3Abacus
$1.5
$9
4SAP AI Core
$1.5
$9
5GitHub Copilot
$1.5
$9
6OpenCode Zen
$1.5
$9
7AIHubMix
$1.5
$9
8Requesty
$1.5
$9
9DevPass (LLM Gateway)
$1.5
$9
10Vertex
$1.5
$9
11Neon
$1.5
$9
12Pioneer
$1.5
$9
13Venice AI
$1.55
$9.45
14Xpersona
$1.55
$12.2
15Cortecs
$1.649
$9.899

比较该模型在不同 API 供应商之间的定价。

外部链接