Google2B (Efficient)~4 GB VRAM at Q4 (QAT)

Gemma 4 E2B — VRAM, speed & local setup

Written by Jakub Rusinowski · Last updated

Gemma 4's most compact variant — the 'Efficient 2B' (~5.1B total params, ~2.3B effective compute via per-layer embeddings) fits in ~5 GB at Q4 and runs on phones, tablets, or any GPU with 6 GB VRAM. Supports text + image input. Outperforms Gemma 3 4B on reasoning and instruction following. Best for offline assistants, edge applications, and Raspberry Pi-style deployments.

Gemma 4 E2B needs about 4 GB of VRAM at Q4 (QAT) — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.

Logic68
Creative63
Coding62

VRAM and speed by quantization

Quoted against NVIDIA RTX 4090 (24 GB). Weights plus overhead only: this model's architecture is unpublished, so no KV cache is added.

QuantVRAMSpeed (est.)Fit
Q2_K
2.63 bpw
2.5 GB~195 tok/sFits
Q3_K_M
3.41 bpw
3 GB~173 tok/sFits
Q4_K_M
4.83 bpw
3.9 GB~143 tok/sFits
Q5_K_M
5.67 bpw
4.4 GB~130 tok/sFits
Q6_K
6.56 bpw
5 GB~118 tok/sFits
Q8_0
8.50 bpw
6.2 GB~99 tok/sFits
F16
16.00 bpw
11 GB~61 tok/sFits

Black marker = usable memory on the NVIDIA RTX 4090 (24 GB). Estimates from the memory-bandwidth roofline on the methodology page. Gemma 4 E2B VRAM calculator →

Get Gemma 4 E2B running

The cheapest catalogued GPU that runs Gemma 4 E2B is the Intel Arc B570 (10 GB).

Buy this hardware Intel Arc B570 10GB — 10 GB VRAM · 150 W board powerDeploy in the cloud now RTX 4090 on RunPod — from $0.34/hr · rate checked 2026-07

or compare on Vast.ai from $0.35/hr (typical low · varies)

As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.

Affiliate disclosure: Some links on this page are affiliate links — if you buy through them, LLM Configurator may earn a commission at no extra cost to you. As an Amazon Associate, LLM Configurator earns from qualifying purchases.
Intel Arc B570 10GB
10 GB VRAM · 150 W board power
2026 prices are volatile — check the current listing.

How to run Gemma 4 E2B

Install Ollama, then run:

ollama run gemma4:e2b
Weights on Hugging Face: google/gemma-4-E2B-it ↗

Specifications

Verified — Checked against the primary source — the model card or the vendor spec page — and corroborated by a second independent source.

Parameters
5.1 Billion (~2.3B effective)
Context window
128,000
Architecture
Dense Transformer + Vision Encoder (per-layer embeddings)
Provider
Google
Licence
Apache 2.0
Practical deployment
Edge device (phone, SBC)
Specified at
Q4 (QAT)
System RAM
8 GB
Record updated
2026-03-31
LicenceApache-2.0Commercial use permitted

Commercial use permitted. No usage restrictions beyond attribution.

Quality and use cases

Scores as published by the model’s authors or an independent evaluator — quality, not throughput, and not measured by us.

Best formobileedge devicesoffline chatlow vramimage understanding
BenchmarkScoreProvenance
MMLU66.4 / 100 %reported
HumanEval58.2 / 100 %reported

Can I run Gemma 4 E2B on my GPU?

Other Gemma 4 sizes

Gemma 4 E2B — frequently asked questions

How much VRAM does Gemma 4 E2B need?

About 4 GB at Q4 (QAT) — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.

Does Gemma 4 E2B run on an RTX 4090 (24 GB)?

Yes. Gemma 4 E2B needs about 4 GB at Q4 (QAT), inside a 24 GB card, at an estimated 143 tokens/sec.

How do I run Gemma 4 E2B locally?

Install Ollama and run `ollama run gemma4:e2b`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.

What other sizes does Gemma 4 come in?

Gemma 4 E2B (4 GB), Gemma 4 E4B (6 GB), Gemma 4 26B-A4B (16 GB), Gemma 4 31B (20 GB), Gemma 4 12B (Unified) (8 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.