跳转到主要内容

Gemma 3 27B Instruct

GoogleGemma开源权重Gemma · 商用许可

描述

Gemma 3 27B is a 27-billion-parameter vision-language model from Google, handling text and image input and generating text output. It features a 128K context window, multilingual support, and open weights. Suitable for complex question answering, summarization, reasoning, and image understanding tasks.

发布日期
2025-03-12
参数规模
27.0B
上下文长度
131K
支持模态
image, text

能力雷达图

60
general
60
coding
80
reasoning
34
science估算
74
agents
70
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
多模态榜101
43.0
LS

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Chat

IFEvalGoogle Research (2023)90.4%自报

Factuality

SimpleQA10.0%自报

General

Global-MMLU-Lite75.1%自报

Language

MMLU-Pro67.5%自报
WMT24++53.4%自报
ECLeKTic16.7%自报

Long Context

MRCR v2 (8-needle)13.5%自报

Math

GSM8k95.9%自报
MATH89.0%自报
MathVista-Mini67.6%自报
HiddenMath60.3%自报

Reasoning

HumanEvalOpenAI (2021)87.8%自报
BIG-Bench Hard87.6%自报
Natural2Code84.5%自报
ChartQAMasry et al. (2022)78.0%自报
FACTS Grounding74.9%自报
MBPP0.74 / 100自报
Bird-SQL (dev)54.4%自报
GPQANYU + Cohere + Anthropic (2023)42.4%自报
LiveCodeBench29.7%自报
BIG-Bench Extra Hard19.3%自报

Vision

DocVQADocVQA (2020)86.6%自报
AI2D84.5%自报
VQAv2 (val)71.0%自报
InfoVQA70.6%自报
TextVQA65.1%自报
MMMU (val)64.9%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))
Chat
90
Instruction Following
90
Structured Output
90
Math
80
Image To Text
70
Legal
70
Multimodal
70
Finance
70
Grounding
70
Healthcare
70
Vision
70
Language
60
Reasoning
60
General
60
Code
60
Physics
40
Factuality
40
Biology
40
Chemistry
40
Long Context
10

定价

输入价格免费
输出价格免费
混合价格(3:1)$0.1 / 1M tokens
缓存读取价格$0.04 / 1M tokens

速度

暂无速度数据

供应商价格排行

供应商价格排行

12 个供应商

最便宜: Kilo Gateway最贵: STACKIT
供应商输入输出
1Kilo Gateway最便宜
$0.064
$0.128
2DeepInfra
$0.08
$0.16
3OpenRouter
$0.08
$0.35
4Hugging Face
$0.08
$0.16
5Deep Infra
$0.08
$0.16
6Merge Gateway
$0.08
$0.45
7AnonRouter
$0.092
$0.184
8Nebius Token Factory
$0.1
$0.3
9Novita AI
$0.119
$0.2
10Prime Inference
$0.119
$0.3
11AI/ML API
$0.1587
$0.48668
12STACKIT
$0.53
$0.76

比较该模型在不同 API 供应商之间的定价。

外部链接