jina embeddings v5 omni nano retrieval needs about 1.0 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 14 quants, from 0.1 GB (IQ1_S) to 0.4 GB (F16).
Multimodal embedding model for text, image, video, and audio
jina-embeddings-v5-omni-nano-retrieval is a GGUF model that produces 768‑dimensional embeddings. It accepts text, images, video, and audio and aligns them in a shared vector space. The model is optimized for retrieval tasks and can be used with llama.cpp and multimodal patches.
Good at
- Retrieval across modalities
- Compact 0.212B parameter size
- Low VRAM footprint
Limits
- No vision support without mmproj
- Requires 8k context for estimates
- CC‑BY‑NC‑4.0 license
Best for
- Cross‑modal search engines
- Embedding‑based recommendation
- Multimodal indexing
Who it’s for. Developers building retrieval systems that need to index text and query with images, video, or audio without reindexing. Suitable for projects that require a small, efficient embedding model.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 0.1 GB | 0.9 GB | Fits | Get file → | |
| IQ1_M | 0.1 GB | 0.9 GB | Fits | Get file → | |
| IQ2_XXS | 0.1 GB | 0.9 GB | Fits | Get file → | |
| IQ2_M | 0.1 GB | 0.9 GB | Fits | Get file → | |
| Q2_K | 0.1 GB | 0.9 GB | Fits | Get file → | |
| Q3_K_M | 0.1 GB | 0.9 GB | Fits | Get file → | |
| IQ4_XS | 0.1 GB | 1.0 GB | Fits | Get file → | |
| IQ4_NL | 0.2 GB | 1.0 GB | Fits | Get file → | |
| Q4_K_M | 0.2 GB | 1.0 GB | Fits | Get file → | |
| Q5_K_S | 0.2 GB | 1.0 GB | Fits | Get file → | |
| Q5_K_M | 0.2 GB | 1.0 GB | Fits | Get file → | |
| Q6_K | 0.2 GB | 1.0 GB | Fits | Get file → | |
| Q8_0 | 0.2 GB | 1.0 GB | Fits | Get file → | |
| F16 | 0.4 GB | 1.2 GB | Fits | Get file → |
For 8 GB VRAM, use IQ1_S or IQ1_M (file size 0.1 GB, estimated 0.9 GB memory). If 12 GB VRAM is available, Q4_K_M or Q5_K_S (file size 0.16 GB or 0.17 GB, estimated 1 GB memory) offer better precision. For the highest quality, F16 (file size 0.43 GB, estimated 1.2 GB memory) is recommended when memory permits.
What GPU do you need for jina embeddings v5 omni nano retrieval?
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 F16 (0.4 GB file, about 1.2 GB in use).
Which jina embeddings v5 omni nano retrieval quant should you choose?
Q4_K_M is the usual starting point for jina embeddings v5 omni nano retrieval: it balances file size and quality well. Here it is a 0.2 GB file.
The Q8_0 file is 0.2 GB, about 1.5 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 IQ1_S 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
Frequently asked questions about jina embeddings v5 omni nano retrieval
How much VRAM does jina embeddings v5 omni nano retrieval need?
About 1.0 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 jina embeddings v5 omni nano retrieval on an 8 GB GPU?
Yes, with F16: the 0.4 GB file needs about 1.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of jina embeddings v5 omni nano retrieval GGUF?
It depends on the quant: IQ1_S is 0.1 GB, IQ1_M is 0.1 GB, IQ2_XXS is 0.1 GB, IQ2_M is 0.1 GB, Q2_K is 0.1 GB, Q3_K_M is 0.1 GB and 8 more.
What context length does jina embeddings v5 omni nano retrieval support?
The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 0.3 GB at 8K and 0.3 GB at 8K.
Where can I download jina embeddings v5 omni nano retrieval GGUF?
The files are on Hugging Face in the jinaai/jina-embeddings-v5-omni-nano-retrieval-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 jina embeddings v5 omni nano retrieval use?
Hugging Face lists the license as cc-by-nc-4.0. 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 this model on 8 GB VRAM?
Yes, the IQ1_S and IQ1_M quantizations use about 0.9 GB memory at 8k context, fitting within 8 GB VRAM.
Does the model support image inputs?
Image inputs are supported when loading the appropriate mmproj file with llama.cpp.
What license applies to this model?
The model is licensed under cc‑by‑nc‑4.0, allowing non‑commercial use with attribution.