Will It Run AI

GoogleGoogle

Gemma 3 12B

当前
2.9M下载量724点赞Mar 2025发布日期131K tokens上下文Gemma许可证70 优秀质量

Gemma 3 12B (12B parameters) requires approximately 14.0 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 17 GB of VRAM.

快速开始

— 复制粘贴即可本地运行

Copy-paste commands to run Gemma 3 12B on your machine.

Run

ollama run gemma3:12b

Quick specs

Parameters12B
Architecturedense
Context131K tokens
Modalitytext
Min RAM4.7 GB
Rec. RAM7.3 GB (Q4_K_M)
LicenseGemma
FamilyGemma
Code Chat

About this model

Gemma 3 12B is Google's mid-range Gemma 3 model with vision capabilities. Offers strong reasoning, code generation, and image understanding balanced with practical resource requirements.

相关模型

你的硬件

检测中...

快速推荐

最佳硬件

Gemma 3 12B 的最佳选择

运行此模型

量化选项

各量化级别的 VRAM 估算

No hardware detected — fit column shows raw VRAM estimates

QuantBitsVRAMQualityFit
Q2_K
2
4.7 GB
Low
Q3_K_S
3
5.9 GB
Low
NVFP4
4
6.7 GB
Medium
Q4_K_M
4
7.3 GB
Medium
Q5_K_M
5
8.6 GB
High
Q6_K
6
9.8 GB
High
Q8_0
8
12.8 GB
Very High
F16
16
24.6 GB
Maximum

Quality benchmarks

Gemma 3 12B benchmark scores

Benchmark verified

Coding

SWE-bench Verified
HumanEval+85.4%
Aider Polyglot
LiveCodeBench24.6%

Reasoning

MMLU-Pro60.6%
GPQA Diamond40.9%
MATH-50083.8%
ARC Challenge

General

Chatbot Arena
IFEval88.9%

Source: official · 2025-03-12

硬件兼容性

全部硬件的适配估算

打开计算器

Computing compatibility...

内存详细分析

Reference: RTX 2060 6GB

Weights7.3 GB
KV Cache4.9 GB
Runtime1.2 GB
Headroom0.6 GB

常见问题

FAQ — Gemma 3 12B

另请参阅