Autor: Jakub Rusinowski · Ostatnia aktualizacja: 31 marca 2026
Model library → Gemma 4 → Gemma 4 E2B
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.
| Parameters | 5.1 Billion (~2.3B effective) |
| Context window | 128,000 |
| Architecture | Dense Transformer + Vision Encoder (per-layer embeddings) |
| Provider | |
| Licence | Apache 2.0 |
| Specified at | Q4 (QAT) |
| System RAM | 8 GB |
| Record updated | 2026-03-31 |
Apache-2.0 — commercial use permitted. Commercial use permitted. No usage restrictions beyond attribution.
Modelled on a reference NVIDIA RTX 4090 (24 GB), with no KV cache (this record has no published architecture). Speed figures are ESTIMATES from the memory-bandwidth roofline described on the methodology page, not benchmarks we ran — rows marked measured come from published or reader-submitted runs. VRAM here includes the KV cache, so it reads higher than the headline figure above, which does not.
| Quant | Weights | VRAM needed | Est. speed | Fit on 24 GB |
|---|---|---|---|---|
| Q2_K | 1.7 GB | 2.5 GB | ~195 tok/s (est.) | Fits comfortably |
| Q3_K_M | 2.2 GB | 3.0 GB | ~173 tok/s (est.) | Fits comfortably |
| Q4_K_M | 3.1 GB | 3.9 GB | ~143 tok/s (est.) | Fits comfortably |
| Q5_K_M | 3.6 GB | 4.4 GB | ~130 tok/s (est.) | Fits comfortably |
| Q6_K | 4.2 GB | 5.0 GB | ~118 tok/s (est.) | Fits comfortably |
| Q8_0 | 5.4 GB | 6.2 GB | ~99 tok/s (est.) | Fits comfortably |
| F16 | 10.2 GB | 11.0 GB | ~61 tok/s (est.) | Fits comfortably |
Want the memory numbers alone, at every quantization level and your own context length? Use the Gemma 4 E2B VRAM calculator.
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.
The cheapest catalogued GPU that runs Gemma 4 E2B is the Intel Arc B570 (10 GB).
Install Ollama, then run:
ollama run gemma4:e2bWeights on Hugging Face: google/gemma-4-E2B-it.
Quality scores as published by the model's authors or an independent evaluator — not throughput, and not measured by us.
| Benchmark | Score | Provenance |
|---|---|---|
| MMLU | 66.4 / 100 % | reported |
| HumanEval | 58.2 / 100 % | reported |
Best for: mobile, edge devices, offline chat, low vram, image understanding.
← All Gemma 4 models | VRAM calculator | Check your own hardware