Llama 3.2 11B Vision

Name: Llama 3.2 11B Vision
Rating: 64 (151 reviews)
Author: Meta

旧版

243.3K下载量1.6K点赞Sep 2024发布日期16K tokens上下文Community许可证36 基础质量

Llama 3.2 11B Vision (11B parameters) requires approximately 10.5 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 13 GB of VRAM.

快速开始

— 复制粘贴即可本地运行

Copy-paste commands to run Llama 3.2 11B Vision on your machine.

Run

ollama run llama3.2-vision:11b

Quick specs

Parameters11B

Architecturevision

Context16K tokens

Modalitytext+vision

Min RAM4.3 GB

Rec. RAM6.7 GB (Q4_K_M)

LicenseCommunity

FamilyLlama Vision

✓ Vision✓ Chat

About this model

Llama 3.2 11B Vision is Meta's multimodal model that processes both text and images. Supports visual question answering, image captioning, and document understanding alongside standard text generation.

快速推荐

最佳性价比B

Intel Arc B580 12GB~$249 — 35 tok/s

最佳综合A

RTX 4080 Super 16GB~$999 — 98 tok/s

最佳硬件

Llama 3.2 11B Vision 的最佳选择

RTX 5080 Laptop 16GBA

16 GB

RTX 4070 Ti Super 16GBA

16 GB

运行此模型

Llama 3.2 11B Vision on RTX 4080 Super 16GB Llama 3.2 11B Vision on RTX 5070 Ti 16GB Llama 3.2 11B Vision on RTX 5080 16GB

量化选项

各量化级别的 VRAM 估算

No hardware detected — fit column shows raw VRAM estimates

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	4.3 GB	Low	—
Q3_K_S	3	5.4 GB	Low	—
NVFP4	4	6.2 GB	Medium	—
Q4_K_M	4	6.7 GB	Medium	—
Q5_K_M	5	7.9 GB	High	—
Q6_K	6	9.0 GB	High	—
Q8_0	8	11.8 GB	Very High	—
F16	16	22.5 GB	Maximum	—

Quality benchmarks

Llama 3.2 11B Vision benchmark scores

Benchmark verified

Reasoning

MMLU-Pro—

GPQA Diamond32.8%

MATH-50051.9%

ARC Challenge—

Source: official · 2024-09-25

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

Weights6.7 GB

KV Cache2.0 GB

Runtime1.2 GB

Headroom0.6 GB

常见问题

FAQ — Llama 3.2 11B Vision

另请参阅

量化指南评分方法打开计算器