Sold by: NVIDIA
Deployed on AWS
Multilingual text question-answering retrieval, transforming textual information into dense vector representations.
Overview
The NVIDIA NeMo Retriever Llama3.2 embedding model is optimized for multilingual and cross-lingual text question-answering retrieval with support for long documents (up to 8192 tokens) and dynamic embedding size (Matryoshka Embeddings). This model was evaluated on 26 languages: English, Arabic, Bengali, Chinese, Czech, Danish, Dutch, Finnish, French, German, Hebrew, Hindi, Hungarian, Indonesian, Italian, Japanese, Korean, Norwegian, Persian, Polish, Portuguese, Russian, Spanish, Swedish, Thai, and Turkish.
In addition to enabling multilingual and cross-lingual question-answering retrieval, this model reduces the data storage footprint by 35x through dynamic embedding sizing and support for longer token length, making it feasible to handle large-scale datasets efficiently.
For additional information please contact NVIDIA: https://www.nvidia.com/en-us/data-center/lp/aws-marketplace-offer
This model is ready for commercial use.
Highlights
- The NeMo Retriever Llama3.2 embedding model is most suitable for users who want to build a multilingual question-and-answer application over a large text corpus, leveraging the latest dense retrieval technologies.
- NVIDIA NIM, a part of the [NVIDIA AI Enterprise](https://www.nvidia.com/en-us/data-center/products/ai-enterprise/) software platform available on the [AWS Marketplace](https://aws.amazon.com/marketplace/pp/prodview-ozgjkov6vq3l6), is a set of easy-to-use microservices designed for secure, reliable deployment of high performance AI model inferencing.
Details
Sold by
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
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.g5.2xlarge Inference (Batch) Recommended | Model inference on the ml.g5.2xlarge instance type, batch mode | $1.00 |
ml.g5.2xlarge Inference (Real-Time) Recommended | Model inference on the ml.g5.2xlarge instance type, real-time mode | $1.00 |
ml.g5.xlarge Inference (Batch) | Model inference on the ml.g5.xlarge instance type, batch mode | $1.00 |
ml.g5.12xlarge Inference (Batch) | Model inference on the ml.g5.12xlarge instance type, batch mode | $4.00 |
ml.g5.8xlarge Inference (Batch) | Model inference on the ml.g5.8xlarge instance type, batch mode | $1.00 |
ml.g5.4xlarge Inference (Batch) | Model inference on the ml.g5.4xlarge instance type, batch mode | $1.00 |
ml.g5.48xlarge Inference (Batch) | Model inference on the ml.g5.48xlarge instance type, batch mode | $8.00 |
ml.g5.16xlarge Inference (Batch) | Model inference on the ml.g5.16xlarge instance type, batch mode | $1.00 |
ml.g5.24xlarge Inference (Batch) | Model inference on the ml.g5.24xlarge instance type, batch mode | $4.00 |
ml.g6.16xlarge Inference (Real-Time) | Model inference on the ml.g6.16xlarge instance type, real-time mode | $1.00 |
Dimensions summary
You pay by the host hour for running this embedding model on a SageMaker instance. Your cost depends on two choices. First, pick an inference mode: batch mode processes data in bulk, while real-time mode serves live requests. Batch mode runs on eight g5 instance sizes. Real-time mode runs across g5, g6, g6e, p4d, p4de, and p5 instance families. Larger instances add more GPU capacity and cost more per hour. You pick the instance size that matches your workload. Billing accrues only for the hours each instance runs.
Top-of-mind questions for buyers
One host hour is one hour that a single SageMaker instance runs the embedding model. The rate matches the instance type you select. Billing accrues by the hour while the instance is active. If you run several instances at once, each one accrues host hours separately.
Software charges accrue only while an instance runs and hosts the model. A stopped instance stops accruing host-hour software charges. Underlying AWS infrastructure fees, such as storage, may still apply separately. The model meters running time, not request volume.
Both bill by the host hour on the instance you choose. Batch mode runs on g5 instances and processes data in bulk jobs. Real-time mode runs on g5, g6, g6e, p4d, p4de, and p5 instances and serves live requests. Pick the mode that matches how you send work.
docs.nvidia.com
Helpful?
Vendor refund policy
No refund
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
Added the NIM_SERVED_MODEL_NAME environment variable. Updated the LangChain Playbook to use the Llama-3.2-NV-EmbedQA-1B-v2 NIM.
Additional details
Inputs
- Summary
The model accepts JSON requests that specifies the input text to be embedded.
{ "input": ["Hello world"], "model": "nvidia/llama-3.2-nv-embedqa-1b-v2", "input_type": "query" }'
- Input MIME type
- application/json
https://docs.api.nvidia.com/nim/reference/nvidia-llama-3_2-nv-embedqa-1b-v2-infer
https://docs.api.nvidia.com/nim/reference/nvidia-llama-3_2-nv-embedqa-1b-v2-infer
Input data descriptions
The following table describes supported input data fields for real-time inference and batch transform.
Field name | Description | Constraints | Required |
|---|---|---|---|
input | Input text to embed. Max length is 512 tokens. | Default value: ""
Type: FreeText | No |
model | ID of the embedding model. | Type: FreeText | Yes |
input_type | passage is used when generating embeddings during indexing. query is used when generating embeddings during querying. | Default value: passage
Type: Categorical
Allowed values: passage, query | No |
encoding_format | The format to return the embeddings in.
| Default value: float
Type: Categorical
Allowed values: float, base64 | No |
truncate | Specifies how inputs longer than the maximum token length of the model are handled. | Default value: none
Type: Categorical
Allowed values: NONE, START, END | No |
Resources
Vendor resources
Support
Vendor support
Free support via NVIDIA NIM Developer Forum: https://forums.developer.nvidia.com/c/ai-data-science/nvidia-nim/
Global enterprise support with NVIDIA AI Enterprise subscription:
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.
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.