Store › model › Embeddings
jina-embeddings-v5-omni-nano-retrieval-GGUF
by jinaai · source Hugging Face · updated 2026-05-12
cc-by-nc-4.0150 MB~1 GB RAMsource aliveunlabeled
jina-embeddings-v5-omni-nano-retrieval-GGUF: Retrieval-Targeted Omni Embedding (Nano) — GGUF
Add to LogiShell Open in the web IDE
The button opens LogiShell with this card; nothing installs from a link by itself. Inside the app the install goes through lsh models install hf:jinaai/jina-embeddings-v5-omni-nano-retrieval-GGUF and its progress lives in the Resource Center.
Source and license
- Source: https://huggingface.co/jinaai/jina-embeddings-v5-omni-nano-retrieval-GGUF
- License: cc-by-nc-4.0 (restricted license: terms at the source apply)
- Requirements: about 1 GB of RAM, 150 MB on disk (estimate: suggested file size × 1.15 + 0.5 GB; a real measurement comes with lsh models). Runs with llama.cpp.
- Tags:
llama.cppggufembeddingqwen3jina-embeddings-v5feature-extractionsentence-transformersmultimodalvisionaudiovideovllmllama-cppimage-feature-extractionaudio-feature-extractionvideo-feature-extraction
Numbers
- 1,829 downloads on Hugging Face
- 1 likes
- license cc-by-nc-4.0
- 0.1 GB for jina-embeddings-v5-omni-nano-retrieval-Q4_K_M.gguf
- 130,822 stars on ggml-org/llama.cpp
- 2,517 open issues and PRs
- last release v0.6.0 on 2026-10-05
- 261,158 npm downloads a week for node-llama-cpp
- latest node-llama-cpp@3.22.1
Numbers as of 2026-10-11 15:33 UTC, from the source API.
Summary
Summary not ready yet: the numbers are here, the text is not. It is written by the collector through the LogiShell model facade when a provider key is present.
Reviews
No reviews yet. Reviews are written inside LogiShell: open this card in the app.
Files
| file | quant | size |
|---|---|---|
jina-embeddings-v5-omni-nano-retrieval-F16.gguf | F16 | 411 MB |
jina-embeddings-v5-omni-nano-retrieval-IQ1_M.gguf | IQ1_M | 97 MB |
jina-embeddings-v5-omni-nano-retrieval-IQ1_S.gguf | IQ1_S | 95 MB |
jina-embeddings-v5-omni-nano-retrieval-IQ2_M.gguf | IQ2_M | 108 MB |
jina-embeddings-v5-omni-nano-retrieval-IQ2_XXS.gguf | IQ2_XXS | 101 MB |
jina-embeddings-v5-omni-nano-retrieval-IQ4_NL.gguf | IQ4_NL | 145 MB |
jina-embeddings-v5-omni-nano-retrieval-IQ4_XS.gguf | IQ4_XS | 142 MB |
jina-embeddings-v5-omni-nano-retrieval-Q2_K.gguf | Q2_K | 124 MB |
jina-embeddings-v5-omni-nano-retrieval-Q3_K_M.gguf | Q3_K_M | 137 MB |
jina-embeddings-v5-omni-nano-retrieval-Q4_K_M.gguf | Q4_K_M | 150 MB |
jina-embeddings-v5-omni-nano-retrieval-Q5_K_M.gguf | Q5_K_M | 161 MB |
jina-embeddings-v5-omni-nano-retrieval-Q5_K_S.gguf | Q5_K_S | 159 MB |
jina-embeddings-v5-omni-nano-retrieval-Q6_K.gguf | Q6_K | 173 MB |
jina-embeddings-v5-omni-nano-retrieval-Q8_0.gguf | Q8_0 | 222 MB |
jina-embeddings-v5-omni-nano-retrieval-audio-mmproj-F16.gguf | F16 | 1.2 GB |
jina-embeddings-v5-omni-nano-retrieval-vision-mmproj-F16.gguf | F16 | 194 MB |
From the source README
jina-embeddings-v5-omni-nano-retrieval-GGUF: Retrieval-Targeted Omni Embedding (Nano) — GGUF
ArXiv | Blog
*Average score vs. parameter count across image (MIEB-Lite), video (MMEB-V), and audio (MAEB) benchmarks — `jina-v5-omni-nano` and `jina-v5-omni-small` define the open-weight frontier (Table 1 in the ArXiv report).*
Model Overview
GGUF + multimodal-projector build of
`jinaai/jina-embeddings-v5-omni-nano-retrieval` for
`llama.cpp`. Accepts text,
images, video, and audio and produces 768-dim embeddings in the
same vector space as the torch reference and as
`jinaai/jina-embeddings-v5-text-nano-retrieval` at the same task —
index with text and query with any modality, no reindexing. For higher performance at a larger size, see `jinaai/jina-embeddings-v5-omni-small-retrieval-GGUF`.
This is the retrieval-targeted variant of the `jina-embeddings-v5-omni-nano` GGUF family. The umbrella with all GGUF variants and cross-repo benchmarks is `jina-ai/jina-embeddings-v5-omni-gguf`.
| Feature | Value |
| --- | --- |
| Parameters | ~0.95B (text + vision + audio towers) |
| Embedding Dimension | 768 |
| Supported Tasks | `retrieval` |
| Max Sequence Length | 8192 |
| Pooling Strategy | Last-token |
| Supported Inputs | text, image, video, audio |
| Supported File Types | images: `.jpg`, `.jpeg`, `.png`, `.gif`, `.webp`, `.bmp`, `.tif`, `.tiff`, `.avif`, `.heic`, `.svg`; video: `.mp4`, `.avi`, `.mov`…
Source: https://huggingface.co/jinaai/jina-embeddings-v5-omni-nano-retrieval-GGUF
Card id model:hf:jinaai/jina-embeddings-v5-omni-nano-retrieval-GGUF · collected 2026-10-11 15:33 UTC · JSON