Sold by: AI21 Labs
Deployed on AWS
Optimized for low-latency processing of long prompts, enabling fast analysis of lengthy documents and data.
Overview
Jamba 1.5 Mini is the first of its kind hybrid Mamba-Transformer architecture at a production grade level offering unmatched efficiency. With an unprecedented context window length (256K) in a smaller model, it offers high quality output for tasks needing large input context & low latency, at a competitive price point.
Highlights
- With a 256K effective long context window, Jamba 1.5 models lead the NVIDIA RULER benchmark—the standard for measuring effective context windows across practical tasks, improving the output quality of key enterprise workflows, such as lengthy document summarization and multi-document analysis.
- 2.5X faster than leading models in its size class on long context and fastest across all context lengths.
- Supports function calling/tool use, structured output (JSON), and grounded generation with citation mode and documents API.
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
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 |
|---|---|---|
ml.m5.large Inference (Batch) Recommended | Model inference on the ml.m5.large instance type, batch mode | $0.00/host/hour |
inference.count.m.i.c Inference Pricing | inference.count.m.i.c Inference Pricing | $30.00/request |
Dimensions summary
You pay for this model based on usage, with two separate dimensions. The first charges by the hour when you run batch inference on the ml.m5.large instance type. You pay for each hour that instance stays active. The second charges per inference request, so your cost tracks the number of requests you send. These dimensions cover different ways to consume the model. Your total depends on how long you run the instance and how many requests you process.
Top-of-mind questions for buyers
You pay for each hour the ml.m5.large instance stays active in batch mode. The clock runs while the instance is provisioned and processing. Stopping the instance ends the hourly software charge, though underlying AWS infrastructure fees may still apply for stored data.
Both dimensions bill independently and can appear together. The ml.m5.large charge tracks how long the batch instance runs. The inference request charge tracks how many requests you send. Running longer raises the hourly total; sending more requests raises the request total. They add up on the same invoice.
Batch inference on ml.m5.large processes grouped inputs while the instance runs, so cost tracks runtime. Per-request billing charges each inference call individually, so cost tracks call volume. The model handles long documents up to a 256K context window under both consumption methods.
www.ai21.com
Helpful?
Vendor refund policy
No refunds.
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
Model release
Additional details
Inputs
- Summary
Jamba 1.5 Mini is the first of its kind hybrid Mamba-Transformer architecture at a production grade level offering unmatched efficiency. With an unprecedented context window length (256K) in a smaller model, it offers high quality output for tasks needing large input context & low latency, at a competitive price point.
- Input MIME type
- application/json
{
"messages":[
{
"role":"user",
"content":"I need help with your product. Can you please assist?"
}
],
"temperature":1,
"top_p":1,
"n":1,
"stop":"\n"
}
https://docs.ai21.com/reference/jamba-15-api-ref#request-parameters
Input data descriptions
The following table describes supported input data fields for real-time inference and batch transform.
Field name | Description | Constraints | Required |
|---|---|---|---|
messages | A list of message objects for the conversation history, which can include system, user, assistant, and tool messages.
See https://docs.ai21.com/reference/jamba-15-api-ref#request-parameters for full description. | Type: FreeText | Yes |
temperature | Controls randomness. allowed range [0,2] | Default value: 1
Type: Continuous
Minimum: 0
Maximum: 2 | No |
top_p | Controls diversity via nucleus sampling. | Default value: 1
Type: Continuous
Minimum: 0
Maximum: 1 | No |
n | Number of completions to generate for each prompt. | Default value: 1
Type: Integer
Minimum: 1
Maximum: 16 | No |
stop | Sequences where the API will stop generating further tokens. Up to 4 sequences | Default value: []
Type: FreeText | No |
max_tokens | Maximum number of tokens to generate. | Default value: 4096
Type: Integer | No |
response_format | Setting to { "type": "json_object" } enables JSON mode, which guarantees the message the model generates is valid JSON | Default value: null
Type: FreeText | No |
tools | A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported.
See https://docs.ai21.com/reference/jamba-15-api-ref#request-parameters for full description. | Default value: []
Type: FreeText | No |
documents | A list of relevant documents the model can ground its responses on, if the user explicitly says so in the prompt. Essentially acts as an extension to the prompt, with the ability to add metadata. each document is a dictionary.
See https://docs.ai21.com/reference/jamba-15-api-ref#request-parameters for full description. | Default value: []
Type: FreeText | No |
Resources
Vendor resources
Support
Vendor support
Talk to our experts: support-amazon@ai21.com
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
Excels at complex reasoning tasks across all prompt lengths, making it ideal for applications that require high quality outputs.
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.