How much VRAM does Llama-3.2-11B-Vision-Instruct need?

Q4_K_M is exactly 5,963,057,216 bytes (5.55 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

How large is Llama-3.2-11B-Vision-Instruct's KV cache?

5.00 GiB at 32K context with an f16 cache, computed per layer from the model's architecture. Quantizing the cache to q8_0 roughly halves it.

meta-llama · text

Llama-3.2-11B-Vision-Instruct

meta-llama/Llama-3.2-11B-Vision-Instruct

Llama-3.2-11B-Vision-Instruct at Q4_K_M is exactly 5,963,057,216 bytes (5.55 GiB / 5.96 GB) — an effective 4.471 bits per weight, not the nominal 4. Its KV cache at 32K is 5.00 GiB.

From the file· summed from 1 file(s)From the file· KV from mirror (mirror:unsloth/Llama-3.2-11B-Vision-Instruct)

Parameters

10.7B

Architecture

mllama

40 layers

Context

131,072

native (config.json)

License

llama3.2

Shipped quantizations

● exact bytes, summed from published files

Quant	Size●	Exact bytes●	Effective bpw●	Tensors●	Publisher
Q4_K_M	5.55 GiB	5,963,057,216	4.471	—	leafspark
Q4_K_M	5.55 GiB	5,963,057,248	4.471	—	pbatra
Q4_K_S	7.01 GiB	7,526,840,896	5.643	—	EnlistedGhost
Q4_K_M	7.28 GiB	7,816,569,408	5.861	—	EnlistedGhost
Q8_0	9.68 GiB	10,394,970,176	7.794	—	leafspark
Q8_0	9.68 GiB	10,394,970,208	7.794	—	pbatra
Q8_0	11.43 GiB	12,272,802,368	9.201	—	EnlistedGhost
F16	18.22 GiB	19,558,899,776	14.664	—	leafspark
F16	20.02 GiB	21,498,171,904	16.118	—	EnlistedGhost

KV cache by context

computed per layer

Context	KV cache (f16)●	Flat formula	Overstated by	Full / windowed / recurrent
4,096	0.63 GiB	0.63 GiB	—	40 / 0 / 0
8,192	1.25 GiB	1.25 GiB	—	40 / 0 / 0
16,384	2.50 GiB	2.50 GiB	—	40 / 0 / 0
32,768	5.00 GiB	5.00 GiB	—	40 / 0 / 0
65,536	10.00 GiB	10.00 GiB	—	40 / 0 / 0
131,072	20.00 GiB	20.00 GiB	—	40 / 0 / 0

Compare with

same modality, comparable size

vs Qwen3-Coder-30B-A3B-Instruct vs Qwen3.6-27B vs Qwen3.8-27B vs gemma-4-12B-it-qat-q4_0-unquantized vs gemma-4-E4B-it vs Qwen3-30B-A3B-Thinking-2507 All alternatives →Will it fit on my card? →

Why other calculators give a different number

A parameters × bits ÷ 8 estimate puts Q4_K_M at roughly 5.59 GiB. The real file is 5.55 GiB, because a quantization is a mixture and some tensors are always kept at higher precision.

Architecture

from mirror:unsloth/Llama-3.2-11B-Vision-Instruct

Layers

Attention heads

KV heads

Head dim

128

Hidden size

4096

Vocab

128,256

Sliding window

none

SWA period

—

MLA

Experts

—

Experts per token

—

use_sliding_window

—

Questions people ask

How much VRAM does Llama-3.2-11B-Vision-Instruct need?: Q4_K_M is exactly 5,963,057,216 bytes (5.55 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
How large is Llama-3.2-11B-Vision-Instruct's KV cache?: 5.00 GiB at 32K context with an f16 cache, computed per layer. Quantizing the cache to q8_0 roughly halves it, which is often the difference between a context length fitting and not.
Which quantization of Llama-3.2-11B-Vision-Instruct should I use?: Q4_K_M is the usual default. Pick the largest quantization that fits your card at the context you actually need — the table above gives exact sizes for every one published.