GE

Gemma 4 E2B IT QAT

GemmaGGUF by pleasenUpdated 2026-10-17
Download on Hugging Face
Parameters4.63B
Context128K
Layers4
Architecturegemma4-assistant
Smallest quant0.1 GB
Quick answer

Gemma 4 E2B IT QAT needs about 0.6 GB of memory at Q4_0 (a 0.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 0.1 GB (Q4_0) to 2.6 GB (Q4_K_XL). Longer context needs more: at 32K tokens the KV cache alone adds about 0.3 GB.

Gemma 4 E2B IT QAT – 4.63B parameter model optimized for mobile and resource‑constrained deployment

Gemma 4 E2B IT QAT is a 4.63 billion parameter assistant model built on the gemma4‑assistant architecture. It supports vision tasks and is optimized for low‑memory environments through targeted quantization and architectural pruning.

Good at

  • Vision inference on mobile devices
  • Low‑memory deployment
  • Fast inference with Q4 quantization

Limits

  • No explicit license information
  • No support for audio encoders
  • Limited to 131072 context tokens

Best for

  • On‑device image captioning
  • Edge‑AI vision assistants
  • Low‑budget inference pipelines

Who it’s for. Developers building lightweight vision‑enabled assistants for mobile or embedded systems. Ideal for projects requiring a small footprint and fast inference without sacrificing visual understanding.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_00.1 GB 0.6 GB
Fits Get file →
Q4_K_XL2.6 GB 3.2 GB
Fits Get file →

For 8 GB VRAM, the Q4_0 quantization (0.06 GB file, 0.6 GB memory at 8 k context) is ideal. If more memory is available, Q4_K_XL (2.62 GB file, 3.2 GB memory) offers higher precision. Both are suitable for 8 k context usage.

What GPU do you need for Gemma 4 E2B IT QAT?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is Q4_K_XL (2.6 GB file, about 3.2 GB in use).

Which Gemma 4 E2B IT QAT quant should you choose?

Q4_0 (0.1 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturegemma4-assistant
Parameters4.63B
Layers4
Trained context131,072 tokens
KV cache (fp16)8 KB per token

Source

Quantized bypleasen
FamilyGemma
Downloads (HF)1,100
Last updated2026-10-17

Frequently asked questions about Gemma 4 E2B IT QAT

How much VRAM does Gemma 4 E2B IT QAT need?

About 0.6 GB at Q4_0 with an 8K context window. That is the 0.1 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Gemma 4 E2B IT QAT on an 8 GB GPU?

Yes, with Q4_K_XL: the 2.6 GB file needs about 3.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Gemma 4 E2B IT QAT GGUF?

It depends on the quant: Q4_0 is 0.1 GB, Q4_K_XL is 2.6 GB.

What context length does Gemma 4 E2B IT QAT support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.1 GB at 8K and 0.3 GB at 32K.

Where can I download Gemma 4 E2B IT QAT GGUF?

The files are on Hugging Face in the pleasen/Gemma-4-E2B-IT-QAT repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Can I run this model on 8 GB VRAM?

Yes, the Q4_0 quantization requires only 0.6 GB memory at 8 k context, well within an 8 GB VRAM budget.

Does it support vision tasks?

Yes, the model includes vision layers and is optimized for visual inference.

What license does it use?

The model is licensed under the gemma license.

Similar models

Browse all →

More from pleasen

All by pleasen →

More Gemma models

All Gemma →

Other small models

All small models →