Gemma 3 12B Instruct
GoogleGemma开源权重Gemma · 商用许可
描述
Gemma 3 12B is a 12-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 question answering, summarization, reasoning, and image understanding tasks.
发布日期
2025-03-12
参数规模
12.0B
上下文长度
131K
支持模态
image, text
能力雷达图
22
general
10
coding
31
reasoning
23
science
25
agents
80
multimodal
排行榜排名
基准测试分数 (LLM Stats)
(LLM Stats (zeroeval))Biology
GPQANYU + Cohere + Anthropic (2023)
40.9%自报
Code
HumanEvalOpenAI (2021)
85.4%自报
LiveCodeBench
24.6%自报
Factuality
FACTS Grounding
75.8%自报
SimpleQA
6.3%自报
Finance
MMLU-Pro
60.6%自报
General
IFEvalGoogle Research (2023)
88.9%自报
Natural2Code
80.7%自报
MBPP
0.73 / 100自报
Global-MMLU-Lite
69.5%自报
MMMU (val)
59.6%自报
BIG-Bench Extra Hard
16.3%自报
Image To Text
DocVQADocVQA (2020)
87.1%自报
VQAv2 (val)
71.6%自报
TextVQA
67.7%自报
Language
BIG-Bench Hard
85.7%自报
WMT24++
51.6%自报
ECLeKTic
10.3%自报
Math
GSM8k
94.4%自报
MATH
83.8%自报
MathVista-Mini
62.9%自报
HiddenMath
54.5%自报
Multimodal
AI2D
84.2%自报
ChartQAMasry et al. (2022)
75.7%自报
InfoVQA
64.9%自报
Reasoning
Bird-SQL (dev)
47.9%自报
AA 评测指数
(Artificial Analysis)Math Index(Artificial Analysis)18.3
Coding Index(Artificial Analysis)5.8
Intelligence Index(Artificial Analysis)5.5
Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))0.9
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))0.6
Ifbench(Google Research (2023))0.4
Gpqa(NYU + Cohere + Anthropic (2023))0.3
Aime(MAA (Mathematical Association of America))0.2
Aime 25(MAA (Mathematical Association of America))0.2
Scicode(UIUC + Argonne National Lab (2024))0.2
Livecodebench(UC Berkeley + MIT + Cornell (2024))0.1
Tau2(Sierra + U Toronto + Vector Institute (2025))0.1
Lcr(Artificial Analysis)0.1
Hle(Center for AI Safety + Scale AI (2025))0.0
Tau Banking0.0
Terminalbench Hard(Stanford × Laude Institute (2026))0.0
Terminalbench V2 10.0
LLM Stats 分类评分
(LLM Stats (zeroeval))Structured Output90
Instruction Following90
Image To Text80
Grounding80
Math70
Multimodal70
Vision70
Legal60
Reasoning60
Finance60
General60
Healthcare60
Code60
Language50
Physics40
Factuality40
Biology40
Chemistry40
定价
输入价格免费
输出价格免费
混合价格(3:1)免费
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
1 个供应商
供应商输入输出
1Neon
$0.15
$0.5
比较该模型在不同 API 供应商之间的定价。