How much VRAM does Gemma-3-4B-VL-it-Gemini-Pro-Heretic-Uncensored-Thinking need?

Q4_K_M is exactly 2,489,895,200 bytes (2.32 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

DavidAU · vision language

Gemma-3-4B-VL-it-Gemini-Pro-Heretic-Uncensored-Thinking

DavidAU/Gemma-3-4B-VL-it-Gemini-Pro-Heretic-Uncensored-Thinking

Gemma-3-4B-VL-it-Gemini-Pro-Heretic-Uncensored-Thinking at Q4_K_M is exactly 2,489,895,200 bytes (2.32 GiB / 2.49 GB) — an effective 4.632 bits per weight, not the nominal 4. Its KV cache at 32K is 0.79 GiB, not the 4.25 GiB a flat formula predicts.

From the file· summed from 1 file(s)From the file· KV per layer

Parameters

4.3B

Architecture

gemma3

34 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
Q2_K	1.61 GiB	1,729,165,600	3.217	—	mradermacher
Q3_K_S	1.80 GiB	1,937,365,280	3.604	—	mradermacher
Q3_K_M	1.95 GiB	2,098,460,960	3.904	—	mradermacher
Q3_K_L	2.08 GiB	2,236,086,560	4.160	—	mradermacher
IQ4_XS	2.12 GiB	2,279,627,040	4.241	—	mradermacher
Q4_K_S	2.21 GiB	2,377,931,040	4.424	—	mradermacher
Q4_K_M	2.32 GiB	2,489,895,200	4.632	—	mradermacher
Q5_K_S	2.57 GiB	2,764,593,440	5.143	—	mradermacher
Q5_K_M	2.64 GiB	2,829,699,360	5.264	—	mradermacher
Q6_K	2.97 GiB	3,190,741,280	5.936	—	mradermacher
Q8_0	3.85 GiB	4,130,403,360	7.684	—	mradermacher
F16	7.23 GiB	7,767,804,960	14.451	—	mradermacher

KV cache by context

computed per layer — this model uses sliding-window attention

Context	KV cache (f16)●	Flat formula	Overstated by	Full / windowed / recurrent
4,096	0.25 GiB	0.53 GiB	2.14×	5 / 29 / 0
8,192	0.33 GiB	1.06 GiB	3.26×	5 / 29 / 0
16,384	0.48 GiB	2.13 GiB	4.40×	5 / 29 / 0
32,768	0.79 GiB	4.25 GiB	5.35×	5 / 29 / 0
65,536	1.42 GiB	8.50 GiB	5.99×	5 / 29 / 0
131,072	2.67 GiB	17.00 GiB	6.37×	5 / 29 / 0

29 of 34 layers cache only a 1,024-token window rather than the full context, on a period of 6. Figures assume the default configuration; --swa-full disables the saving entirely.

Compare with

same modality, comparable size

vs Qwen3.5-9B vs gemma-4-12B-it vs Qwen3.5-4B vs Qwen3.5-0.8B vs gemma-4-E2B-it vs Qwythos-9B-v2 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 2.25 GiB. The real file is 2.32 GiB, because a quantization is a mixture and some tensors are always kept at higher precision. The larger discrepancy is the cache: a flat formula gives 4.25 GiB at 32K context where the real figure is 0.79 GiB, because most of this model's layers cache a fixed window rather than the whole context.

Architecture

from config.json

Layers

Attention heads

KV heads

Head dim

256

Hidden size

2560

Vocab

262,208

Sliding window

1024

SWA period

MLA

Experts

—

Experts per token

—

use_sliding_window

—

Questions people ask

How much VRAM does Gemma-3-4B-VL-it-Gemini-Pro-Heretic-Uncensored-Thinking need?: Q4_K_M is exactly 2,489,895,200 bytes (2.32 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
How large is Gemma-3-4B-VL-it-Gemini-Pro-Heretic-Uncensored-Thinking's KV cache?: 0.79 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 Gemma-3-4B-VL-it-Gemini-Pro-Heretic-Uncensored-Thinking 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.