Sold by: Jina AI
Deployed on AWS
Jina Embeddings v5 Omni Nano is a 1.04B-parameter multimodal embedding model that maps text, images, video, audio, and PDFs into a single shared vector space, with an 8K-token text context, Matryoshka dimensions from 32 to 768, and task-specific LoRA adapters for retrieval, classification, clustering, and text matching, with multilingual support across dozens of languages.
Overview
Jina Embeddings v5 Omni Nano is the compact multimodal member of the latest generation of Jina AI's open-weight embedding family. Built on a EuroBERT-210m text backbone with a SigLIP2 Base vision tower and a Whisper-large-v3 audio tower, its 1.04B parameters map text, images, video, audio, and PDF documents into a single shared embedding space, so you can index any modality and retrieve any other from one vector index. This enables cross-modal search, multimodal RAG, visual document retrieval over PDFs and scans, video moment retrieval, and audio semantic search without separate per-modality pipelines. Text embeddings are identical to jina-embeddings-v5-text-nano, so multimodal content drops into an existing v5-text-nano index with no reindexing. The model handles an 8,192-token text context, supports multilingual text across dozens of languages, and delivers competitive document retrieval at just over a billion parameters, outperforming similarly sized open-weight omni models on the multimodal Pareto frontier. Matryoshka Representation Learning lets you truncate embeddings from 768 down to 32 dimensions without retraining, trading storage cost for marginal recall loss. Four task-specific LoRA adapters (retrieval, text-matching, clustering, and classification) tune the same base model for different downstream workloads at request time.
Highlights
- One shared embedding space for every modality: index text, images, video, audio, and PDFs together and query across them, so a text query can retrieve a video frame, a scanned page, or an audio clip from a single vector index.
- Drop-in compatible with jina-embeddings-v5-text-nano: text embeddings are identical, so you can add multimodal content to an existing v5-text-nano index with no reindexing. At just 1.04B parameters it runs lighter and cheaper than larger omni models while staying competitive on cross-modal retrieval.
- Matryoshka dimensions from 32 to 768 with four task-specific LoRA adapters: truncate embeddings to fit your storage and latency budget, and switch between retrieval, text-matching, clustering, and classification per request from a single deployed model.
Details
Sold by
Categories
Delivery method
Latest version
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Trust Center
Access real-time vendor security and compliance information through their Trust Center powered by Drata or Vanta. Review certifications and security standards before purchase.
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.g5.xlarge Inference (Batch) Recommended | Model inference on the ml.g5.xlarge instance type, batch mode | $2.50 |
ml.g5.xlarge Inference (Real-Time) Recommended | Model inference on the ml.g5.xlarge instance type, real-time mode | $2.50 |
ml.g4dn.xlarge Inference (Batch) | Model inference on the ml.g4dn.xlarge instance type, batch mode | $2.50 |
ml.g4dn.2xlarge Inference (Batch) | Model inference on the ml.g4dn.2xlarge instance type, batch mode | $2.50 |
ml.g4dn.4xlarge Inference (Batch) | Model inference on the ml.g4dn.4xlarge instance type, batch mode | $2.50 |
ml.g4dn.8xlarge Inference (Batch) | Model inference on the ml.g4dn.8xlarge instance type, batch mode | $2.50 |
ml.g4dn.12xlarge Inference (Batch) | Model inference on the ml.g4dn.12xlarge instance type, batch mode | $2.50 |
ml.g4dn.16xlarge Inference (Batch) | Model inference on the ml.g4dn.16xlarge instance type, batch mode | $2.50 |
ml.g6.xlarge Inference (Batch) | Model inference on the ml.g6.xlarge instance type, batch mode | $2.50 |
ml.g5.2xlarge Inference (Real-Time) | Model inference on the ml.g5.2xlarge instance type, real-time mode | $2.50 |
Dimensions summary
You pay by the hour for each host running the model, so cost scales with how long you run inference and which instance you pick. Dimensions split into two modes: batch inference (for processing groups of data at once) and real-time inference (for live requests). Within each mode, you choose a GPU instance size across the g4dn, g5, and g6 families. Larger instance sizes carry more compute, so hourly rates rise as you move up within a family. Batch mode covers smaller g4dn, g5, and g6 sizes; real-time mode spans g5 and g6 sizes.
Top-of-mind questions for buyers
One HostHrs unit is one hour that a chosen instance runs the model. Charges accrue per running host-hour based on the instance type you launch. You keep paying while the endpoint stays running, even without active requests. Stopping or deleting the endpoint stops the software charge. Underlying AWS infrastructure fees are separate.
Both bill per host-hour, but they meter different workload styles. Batch mode runs the model to process grouped data in one pass, then you shut the host down. Real-time mode keeps a host running to answer live requests. Batch suits scheduled bulk jobs; real-time suits continuous serving. You pick the mode by selecting the matching dimension.
You are billed at the hourly rate for whichever instance you run. Moving to a larger size within the g4dn, g5, or g6 family raises the hourly rate because the instance carries more GPU compute. The change is not automatic. You choose the instance size when you deploy, and cost follows that choice.
jina.ai
Helpful?
Vendor refund policy
For support, please visit https://jina.ai/contact-sales or https://www.elastic.co/support .
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
Amazon SageMaker model
An Amazon SageMaker model package is a pre-trained machine learning model ready to use without additional training. Use the model package to create a model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Deploy the model on Amazon SageMaker AI using the following options:
Version release notes
Initial release
Additional details
Inputs
- Summary
The model accepts JSON inputs. Texts must be passed in the following format.
- Input MIME type
- text/csv
https://github.com/jina-ai/jina-sagemaker/blob/main/examples/sample-inference-v5-omni-nano-input.json
https://github.com/jina-ai/jina-sagemaker/blob/main/examples/sample-batch-v5-omni-nano-input.csv
Support
Vendor support
For support, please visit https://jina.ai/contact-sales or
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
Text embedding model (base) for input of size up to 8192 tokens
Jina ColBERT v2 Reranker model for Multilingual Documents
Embedding model for cross-modal and multimodal retrieval for text and image data
New State-of-the-Art Multilingual Embeddings With Task LoRA
Text embedding model (small) for input of size up to 8192 tokens
Customer reviews
No customer reviews yet
Be the first to review this product . We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.