GE

Sakura Gemma4 E4B EmbeddingGemma2 DualMode

GemmaGGUF by webmp3Updated 2026-10-08
Download on Hugging Face
Parameters7.52B
Context128K
Layers42
Architecturegemma4
Smallest quant5.8 GB
Quick answer

Sakura Gemma4 E4B EmbeddingGemma2 DualMode needs about 7.7 GB of memory at Q5_K_M (a 5.8 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 2 quants, from 5.8 GB (Q5_K_M) to 5.8 GB (Q5_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 5.6 GB.

Sakura Gemma4 E4B DualMode – chat and 768‑d embeddings in one GGUF

Sakura Gemma4 E4B DualMode is a Gemma 4 E4B-it model that supports both generation and 768‑dimensional embeddings. It uses a single GGUF file with a tiny adapter and head for embedding mode. The model is a research prototype and not an official Google release.

Good at

  • Chat generation
  • 768‑d embeddings for RAG
  • Bidirectional attention for embeddings

Limits

  • Requires a llama.cpp patch to load
  • No vision or audio support
  • License is gemma (Apache 2.0)

Best for

  • Local chat applications
  • Embedding‑based retrieval
  • RAG pipelines

Who it’s for. Developers needing a single model for both text generation and embedding tasks on local devices. Suitable for laptops and servers with limited storage.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q5_K_M5.8 GB 7.7 GB
Tight Get file →
Q5_K_M5.8 GB 7.7 GB
Tight Get file →

The available GGUF file is Q5_K_M, 5.76 GB, requiring about 7.7 GB of VRAM at 8 k context. This is the only quantized version provided, so use it if your GPU can accommodate the memory estimate.

What GPU do you need for Sakura Gemma4 E4B EmbeddingGemma2 DualMode?

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: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: the best quant that fits is Q5_K_M (5.8 GB file, about 7.7 GB in use).

Which Sakura Gemma4 E4B EmbeddingGemma2 DualMode quant should you choose?

Q5_K_M (5.8 GB) is the middle option in this repository. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturegemma4
Parameters7.52B
Layers42
Trained context131,072 tokens
KV cache (fp16)168 KB per token

Source

Quantized bywebmp3
FamilyGemma
Licensegemma
Downloads (HF)580
Last updated2026-10-08

Frequently asked questions about Sakura Gemma4 E4B EmbeddingGemma2 DualMode

How much VRAM does Sakura Gemma4 E4B EmbeddingGemma2 DualMode need?

About 7.7 GB at Q5_K_M with an 8K context window. That is the 5.8 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 Sakura Gemma4 E4B EmbeddingGemma2 DualMode on an 8 GB GPU?

Yes, with Q5_K_M: the 5.8 GB file needs about 7.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Sakura Gemma4 E4B EmbeddingGemma2 DualMode GGUF?

It depends on the quant: Q5_K_M is 5.8 GB, Q5_K_M is 5.8 GB.

What context length does Sakura Gemma4 E4B EmbeddingGemma2 DualMode support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.4 GB at 8K and 5.6 GB at 32K.

Where can I download Sakura Gemma4 E4B EmbeddingGemma2 DualMode GGUF?

The files are on Hugging Face in the webmp3/Sakura-Gemma4-E4B-EmbeddingGemma2-DualMode-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Sakura Gemma4 E4B EmbeddingGemma2 DualMode use?

Hugging Face lists the license as gemma. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run this model on 8 GB VRAM?

The estimated VRAM usage is 7.7 GB at 8 k context, so it can fit on an 8 GB GPU with some headroom.

Does it support image or audio input?

No, the model only handles text input and output.

Is this an official Google model?

No, it is a community research prototype and not an official Google release.

Other GGUFs of gemma-4-E4B-it

Similar models

Browse all →

More from webmp3

All by webmp3 →

More Gemma models

All Gemma →

Other small models

All small models →