跳转到主要内容

Gemini 3.5 Flash-Lite

GoogleGeminiProprietary

描述

Gemini 3.5 Flash-Lite is Google's low-latency, cost-effective multimodal reasoning model for high-throughput agentic workflows, document processing, data extraction, translation, and classification. It supports text, image, video, audio, and PDF inputs, a 1 million-token context window, and a 65,536-token text output.

发布日期
2026-07-21
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

26
general
48
coding
84
reasoning
56
science
50
agents
80
multimodal

排行榜排名

领域#排名分数来源
代码能力榜140
69.0
AA
通用能力榜249
48.0
AA
数学推理45
74.0
LB
多模态榜64
49.0
LS
推理能力45
60.0
LB
科学能力179
57.0
AA

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Code

MLE-Bench39.2%自报

Long Context

MRCR v2 (8-needle)21.3%自报

Multimodal

OSWorld-Verified74.0%自报

Reasoning

CharXiv-R76.5%自报
SWE-Bench ProPrinceton NLP (2024)54.2%自报
Terminal-Bench 2.154.0%自报

AA 评测指数

(Artificial Analysis)
Gpqa(NYU + Cohere + Anthropic (2023))
83.8
Lcr(Artificial Analysis)
76.0
Terminalbench V2 1
53.6
Coding Index(Artificial Analysis)
49.3
Scicode(UIUC + Argonne National Lab (2024))
41.3
Intelligence Index(Artificial Analysis)
27.6
Hle(Center for AI Safety + Scale AI (2025))
18.8
Tau Banking
17.5

LLM Stats 分类评分

(LLM Stats (zeroeval))
Multimodal
80
Vision
80
General
60
Agents
60
Reasoning
50
Code
50
Tool Calling
50
Long Context
20

定价

输入价格$0.3 / 1M tokens
输出价格$2.5 / 1M tokens
混合价格(3:1)$0.85 / 1M tokens
缓存读取价格$0.03 / 1M tokens

速度

Tokens/秒332.1
首Token延迟6.75s
首回答延迟6.75s

供应商价格排行

供应商价格排行

19 个供应商

最便宜: Google最贵: Venice AI
供应商输入输出
1Google最便宜
$0
$0
2Kilo Gateway
$0.15
$1.25
3NanoGPT
$0.3
$2.5
4Abacus
$0.3
$2.5
5OpenRouter
$0.3
$2.5
6OpenCode Zen
$0.3
$2.5
7Requesty
$0.3
$2.5
8Vercel AI Gateway
$0.3
$2.5
9DevPass (LLM Gateway)
$0.3
$2.5
10Vertex
$0.3
$2.5
11OrcaRouter
$0.3
$2.5
12Merge Gateway
$0.3
$2.5
13Neon
$0.3
$2.5
14Pioneer
$0.3
$2.5
15Ofox
$0.3
$2.5
16Impossibl
$0.3
$2.5
17Eden AI
$0.3
$2.5
18Cortecs
$0.33
$2.749
19Venice AI
$0.375
$3.125

比较该模型在不同 API 供应商之间的定价。

外部链接