Sold by: Multiverse Computing
Deployed on AWS
Reduce inference costs and increase inference speed by using Multiverse Computing's CompactifAI Llama 4 Scout Slim. A 50% compression of the widely known Meta Llama 4 Scout model.
Overview
Reduce inference costs and increase inference speed by using Multiverse Computing's CompactifAI Llama 4 Scout Slim. A 50% compression of the widely known Meta Llama 4 Scout model.
Highlights
- 50% less memory usage
- 25% Faster Inference
- Maintains 97% of the original model intelligence
Details
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.xlarge Inference (Batch) Recommended | Model inference on the ml.g5.xlarge instance type, batch mode | $0.29 |
ml.p5.48xlarge Inference (Real-Time) Recommended | Model inference on the ml.p5.48xlarge instance type, real-time mode | $7.72 |
Dimensions summary
You pay by the hour for model inference, based on the AWS instance type and inference mode you run. The ml.g5.xlarge option runs in batch mode, which processes grouped requests rather than immediate responses. The ml.p5.48xlarge option runs in real-time mode for on-demand responses. These are independent usage-based options, not tiers. Your cost scales with how many host hours you use on each instance. You select the instance and mode that fit your workload, and billing follows actual usage.
Top-of-mind questions for buyers
Each host hour covers running the compressed model on that specific AWS instance type. The ml.g5.xlarge is a smaller single-GPU instance suited to batch processing. The ml.p5.48xlarge is a multi-GPU instance built for real-time inference. You pay for each hour the chosen instance runs.
Billing meters host hours, so charges accrue while the instance is provisioned and running. If you stop or terminate the instance, software host-hour charges stop. Underlying AWS infrastructure fees, such as storage, may still apply separately based on your AWS account setup.
The ml.g5.xlarge batch option processes grouped requests together, which fits scheduled or non-urgent workloads. The ml.p5.48xlarge real-time option handles on-demand responses for interactive use. Each bills independently by host hour. You can run one, the other, or both, depending on your workload.
multiversecomputing.com
Helpful?
Vendor refund policy
Placeholder Refund Policy
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
CompactifAI Llama 4 Scout inference on a vLLM engine
Additional details
Inputs
- Summary
Our model accepts an input compatible with openAI's chat completions endpoint.
- Limitations for input type
- Context Length Limit: `256000` Maximum Output Sequence Length: `32768`
- Input MIME type
- application/json
{
"model": "cai-llama-4-scout-slim",
"messages": [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Tell me about quantum computing."}
]
}
{
"model": "cai-llama-4-scout-slim",
"messages": [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Tell me about quantum computing."}
]
}
Input data descriptions
The following table describes supported input data fields for real-time inference and batch transform.
Field name | Description | Constraints | Required |
|---|---|---|---|
messages | An array of dictionaries, each dictionary must follow this format:
{
"role": "system" | "user" | "assistant",
"content": "The message content"
} | The sum of the tokens of your messages array must not exceed 223232 | Yes |
model | A valid model name: `cai-llama-4-scout-slim` | - | Yes |
Support
Vendor support
Send an email to support@compactif.ai and provide a description of your issue
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
CompactifAI API empowers organizations with ultra-efficient and scalable AI models that slash compute and energy costs, accelerate deployment, and fuel innovation, all without compromising performance or reliability!
Reduce inference costs and increase inference speed by using Multiverse Computing's CompactifAI Llama 3.1 Slim. A 80% compression of the widely known Meta Llama 3.1 8B model.
Reduce inference costs and increase inference speed by using Multiverse Computing's CompactifAI HyperNova 60B. In direct benchmarks against Mistral 3 Large, HyperNova 60B delivers outstanding gains:
Reduce inference costs and increase inference speed by using Multiverse Computing's CompactifAI Mistral Small 3.1 Slim. A 50% compression of the widely known Mistral Small 3.1 model.
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.