EM

xVITA Embed2 zhTW

EmbeddinggemmaGGUF by xCloudinfoUpdated 2026-10-26
Download on Hugging Face
Parameters271M
Context256K
Layers24
Architecturegemma-embedding2
Smallest quant0.1 GB
Quick answer

xVITA Embed2 zhTW needs about 2.3 GB of memory at Q4_K_M (a 0.2 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 7 quants, from 0.1 GB (IQ2_M) to 0.6 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 6.4 GB.

xVITA Embed2 zhTW – 768‑dimensional multimodal embedding model

xVITA Embed2 zhTW is a fine‑tuned EmbeddingGemma 2 model for Traditional Chinese. It maps text, images, video, and audio into a shared 768‑dimensional vector space. The model runs on consumer hardware via llama.cpp.

Good at

  • Multimodal embeddings
  • Chinese text retrieval
  • Low‑latency vector generation

Limits

  • No text generation
  • Requires 8K context window
  • License: Apache‑2.0

Best for

  • Search engines
  • RAG pipelines
  • Multimodal clustering

Who it’s for. Developers building on‑device search or retrieval systems that need unified embeddings for text, images, video, and audio in Traditional Chinese. The model is suitable for mobile or laptop deployments where low memory and latency are important.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_M0.1 GB 2.2 GB
Fits Get file →
IQ4_XS0.2 GB 2.3 GB
Fits Get file →
Q4_K_M0.2 GB 2.3 GB
Fits Get file →
Q5_K_M0.2 GB 2.3 GB
Fits Get file →
Q6_K0.2 GB 2.4 GB
Fits Get file →
Q8_00.3 GB 2.4 GB
Fits Get file →
BF160.6 GB 2.7 GB
Fits Get file →

For 8 GB VRAM, use IQ4_XS or Q4_K_M (file size 0.17 GB or 0.18 GB, memory ~2.3 GB). If you have 12 GB VRAM, Q6_K or Q8_0 (0.25 GB or 0.31 GB) provide higher precision. BF16 offers the highest quality but needs ~2.7 GB memory.

What GPU do you need for xVITA Embed2 zhTW?

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 BF16 (0.6 GB file, about 2.7 GB in use).

Which xVITA Embed2 zhTW quant should you choose?

Q4_K_M is the usual starting point for xVITA Embed2 zhTW: it balances file size and quality well. Here it is a 0.2 GB file.

The Q8_0 file is 0.3 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is IQ2_M at 0.1 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturegemma-embedding2
Parameters271M
Layers24
Trained context262,144 tokens
KV cache (fp16)192 KB per token

Source

Quantized byxCloudinfo
Licenseapache-2.0
Downloads (HF)1,751
Last updated2026-10-26

Frequently asked questions about xVITA Embed2 zhTW

How much VRAM does xVITA Embed2 zhTW need?

About 2.3 GB at Q4_K_M with an 8K context window. That is the 0.2 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 xVITA Embed2 zhTW on an 8 GB GPU?

Yes, with BF16: the 0.6 GB file needs about 2.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of xVITA Embed2 zhTW GGUF?

It depends on the quant: IQ2_M is 0.1 GB, IQ4_XS is 0.2 GB, Q4_K_M is 0.2 GB, Q5_K_M is 0.2 GB, Q6_K is 0.2 GB, Q8_0 is 0.3 GB and 1 more.

What context length does xVITA Embed2 zhTW support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.6 GB at 8K and 6.4 GB at 32K.

Where can I download xVITA Embed2 zhTW GGUF?

The files are on Hugging Face in the xCloudinfo/xVITA-Embed2-zhTW-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 xVITA Embed2 zhTW use?

Hugging Face lists the license as apache-2.0. 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 xVITA Embed2 zhTW on 8 GB VRAM?

Yes, the IQ4_XS and Q4_K_M quantizations fit within 8 GB VRAM, using about 2.3 GB of memory at 8K context.

Does this model support vision?

Yes, it includes a vision encoder and can embed images into the same vector space as text.

Is the model open source?

It is released under the Apache‑2.0 license and can be used freely in commercial or non‑commercial projects.

Other GGUFs of embeddinggemma-2

Similar models

Browse all →

More from xCloudinfo

All by xCloudinfo →

More Embeddinggemma models

All Embeddinggemma →

Other tiny models

All tiny models →