How much VRAM does Llama-3.2-3B need?

Q4_K_M is exactly 2,019,377,632 bytes (1.88 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

How large is Llama-3.2-3B's KV cache?

3.50 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-3.2-3B

meta-llama/Llama-3.2-3B

Llama-3.2-3B at Q4_K_M is exactly 2,019,377,632 bytes (1.88 GiB / 2.02 GB) — an effective 5.028 bits per weight, not the nominal 4. Its KV cache at 32K is 3.50 GiB.

From the file· summed from 1 file(s)From the file· KV from mirror (mirror:unsloth/Llama-3.2-3B)

Parameters

3.2B

Architecture

llama

28 layers

Context

131,072

native (config.json)

License

apache-2.0

Shipped quantizations

● exact bytes, summed from published files

Quant	Size●	Exact bytes●	Effective bpw●	Tensors●	Publisher
TQ1_0	0.86 GiB	926,286,304	2.307	—	osmapi
TQ2_0	0.99 GiB	1,058,406,880	2.635	—	osmapi
Q2_K	1.27 GiB	1,363,935,712	3.396	—	osmapi
Q3_K_S	1.44 GiB	1,542,848,992	3.842	—	osmapi
Q3_K_M	1.57 GiB	1,687,159,264	4.201	—	osmapi
Q3_K_L	1.69 GiB	1,815,347,680	4.520	—	osmapi
Q4_K_S	1.80 GiB	1,928,200,672	4.801	—	osmapi
Q4_K_M	1.88 GiB	2,019,377,632	5.028	—	osmapi
Q5_K_S	2.11 GiB	2,269,512,160	5.651	—	osmapi
Q5_K_M	2.16 GiB	2,322,153,952	5.782	—	osmapi
Q6_K	2.46 GiB	2,643,853,792	6.583	—	osmapi
Q8_0	3.19 GiB	3,421,895,520	8.521	—	ABDALLALSWAITI
F162 shards	11.98 GiB	12,867,376,064	32.041	—	osmapi

KV cache by context

computed per layer

Context	KV cache (f16)●	Flat formula	Overstated by	Full / windowed / recurrent
4,096	0.44 GiB	0.44 GiB	—	28 / 0 / 0
8,192	0.88 GiB	0.88 GiB	—	28 / 0 / 0
16,384	1.75 GiB	1.75 GiB	—	28 / 0 / 0
32,768	3.50 GiB	3.50 GiB	—	28 / 0 / 0
65,536	7.00 GiB	7.00 GiB	—	28 / 0 / 0
131,072	14.00 GiB	14.00 GiB	—	28 / 0 / 0

Compare with

same modality, comparable size

vs Qwen3-4B vs Qwen3-8B vs HyperCLOVAX-SEED-Text-Instruct-1.5B vs Llama-3.2-1B-Instruct vs Qwen3-VL-8B-Instruct-abliterated-v1 vs Llama-3.1-8B-Instruct 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 1.68 GiB. The real file is 1.88 GiB, because a quantization is a mixture and some tensors are always kept at higher precision.

Architecture

from mirror:unsloth/Llama-3.2-3B

Layers

Attention heads

KV heads

Head dim

128

Hidden size

3072

Vocab

128,256

Sliding window

none

SWA period

—

MLA

Experts

—

Experts per token

—

use_sliding_window

—

Questions people ask

How much VRAM does Llama-3.2-3B need?: Q4_K_M is exactly 2,019,377,632 bytes (1.88 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-3.2-3B's KV cache?: 3.50 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-3.2-3B 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.