Sold by: SpadEx Inc
Deployed on AWS
SpadEx Pluggable Agentic Last Mile (PALM) provides a VPC-deployed AMI for secure document understanding workloads at scale. PALM's multi-LLM inference pipeline optimizes token costs and processing accuracy while keeping sensitive data within your VPC.
Overview
Pluggable Agentic Last Mile (PALM) by SpadEx
PALM is a self-contained Amazon Machine Image (AMI) that processes complex document understanding workloads entirely within your Amazon Virtual Private Cloud (VPC). Private documents and sensitive corporate data never leave your controlled cloud environment, giving compliance and security teams confidence that data sovereignty requirements are met by architecture - not just policy.
Secure Multi-Agent Orchestration
At the core of PALM is an advanced three-LLM pipeline that intelligently splits orchestration and local inference to maximize accuracy.
- Pre-computed reasoning paths - Public instructions or shareable templates can be analyzed by external cloud LLMs to generate reasoning and code ahead of time. You can also paste pre-existing reasoning blocks directly into the pipeline, skipping cloud-side LLM calls entirely for zero external API cost on repeat workloads.
- Local agent dispatch - PALM disassembles complex analysis instructions into structured execution plans and dispatches them to dynamically generated local agents running within the AMI.
- Native fallback mode - If external code generation configuration is omitted, PALM runs direct step-by-step local model calls, keeping workflows fully operational without any external model dependencies.
Self-Healing Extractions
PALM includes a built-in quality assurance loop that classifies extraction failures and routes them to the appropriate fix. Format or syntax errors trigger a lightweight reformat sub-agent that corrects constraints without re-scanning the document image.
Tokenomics and Cost Optimization
PALM is engineered to reduce cloud token expenses through multiple mechanisms:
- Pre-computed reasoning paths allow repeat workloads to bypass external API calls entirely.
- Local inference handles private document content without per-token charges from external providers.
- The self-healing retry loop avoids costly full-document re-processing by isolating and fixing only the failed extraction step.
Getting Started
PALM is delivered as a single AMI that you launch within your VPC. To begin evaluating PALM for your document workloads, subscribe through AWS Marketplace and launch the AMI in your target VPC. Contact SpadEx at support@spadexinc.com to schedule a guided architecture walkthrough or request assistance with your initial deployment configuration.
Next Steps
To discuss a pilot deployment, book a discovery call with the SpadEx team by emailing support@spadexinc.com with your use case details and target document types.
Highlights
- Complete data sovereignty by design: Every private document understanding step runs within your Amazon VPC. Private documents and sensitive corporate data never leave your controlled cloud environment. PALM's design ensures data sovereignty through infrastructure isolation rather than policy alone, making it suitable for regulated industries and workloads where strict data-handling requirements apply.
- Token Cost Optimization: PALM's pre-computed reasoning paths let you insert previously generated reasoning blocks directly into the pipeline, bypassing cloud-side LLM orchestration calls entirely. For new workloads, local inference handles all private document content within the AMI, and a native fallback mode ensures full operation even without external code generation - reducing unpredictable token expenses.
- PALM's three-LLM pipeline enhances local model capabilities with context you provide about your document structure and enhances extraction accuracy.
Details
Sold by
Delivery method
Delivery option
64-bit (x86) Amazon Machine Image (AMI)
Latest version
Operating system
AmazonLinux 2023
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 | Cost/hour |
|---|---|
g6e.xlarge Recommended | $1.00 |
g7e.48xlarge | $1.00 |
g6e.16xlarge | $1.00 |
g7e.8xlarge | $1.00 |
g6.xlarge | $1.00 |
g6e.8xlarge | $1.00 |
g5.xlarge | $1.00 |
g6.12xlarge | $1.00 |
g5.12xlarge | $1.00 |
g6.24xlarge | $1.00 |
Vendor refund policy
Refunds are not available at this time for the paid hourly PALM AMI. You may stop instances using the AMI at anytime to avoid additional charges.
Exception: if you are unable to use the AMI due to an outage in the PALM Server you can request a refund for the time when the server was unavailable. Contact support@spadexinc.com with: 1. Account ID and region, 2. Instance ID, 3. Instance launch times, 4. A brief description of the issue. We will process your request within 5 U.S. business days.
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
64-bit (x86) Amazon Machine Image (AMI)
Amazon Machine Image (AMI)
An AMI is a virtual image that provides the information required to launch an instance. Amazon EC2 (Elastic Compute Cloud) instances are virtual servers on which you can run your applications and workloads, offering varying combinations of CPU, memory, storage, and networking resources. You can launch as many instances from as many different AMIs as you need.
Version release notes
This is the initial public release. Please see the user guide for details.
Additional details
Usage instructions
Please see the full PALM user guide at https://www.spadexinc.com/palm-user-guide
After starting an instance using the PALM AMI use the EC2 instance ID (e.g. i-1234abcd5678ef90a) to login to the Web UI. The PALM AMI does not require access keys or access to other AWS services to operate.
The PALM Web UI is accessible on port 7860 and the PALM Service is accessible on port 8080. To access these ports outside of your VPC, allow inbound access from your CIDR IP blocks to ports 7860 and 8080 in the Security Group configured for the instance. Alternately, you may use SSH port forwarding to these ports as described in the full user guide. You must allow outbound traffic in the Security Group.
The PALM Web UI will be accessible at https://<Instance IP>:7860. The UI webservice may take up to 5 minutes to launch after launching the EC2 instance. You may see a placeholder page or a 502 error code while the instance is still starting up. All traffic to the instance is encrypted over TLS using a self-signed certificate within the AMI. This may cause certificate warnings. See the full user guide for more details.
The PALM AMI uses an API hosted by SpadEx Inc to support reference reasoning and external code generation. PALM connects to this API using the EC2 instance ID obtained from the instance metadata when you login to the Web UI. No additional customer information is required. No IAM Roles or Policies are required.
The full user guide describes further how to use the PALM Service API and how to enable logging.
Resources
Vendor resources
Support
Vendor support
Support for PALM AMI
SpadEx provides technical support for the PALM AMI product via email at support@spadexinc.com .
Standard Support
- Channel: Email (support@spadexinc.com )
- Response time: Within 24 hours on U.S. business days
- Scope: Installation assistance, configuration questions, troubleshooting extraction issues, and general product guidance
Enhanced Support Tiers
Higher tiers of support - including priority response times and dedicated assistance - are available with a monthly or annual PALM subscription. Contact support@spadexinc.com for details on enhanced support options.
Getting Help
When submitting a support request, please include:
- Your AWS account ID and region
- The PALM AMI version you are running
- A description of the issue or question
- Any relevant error messages or logs
Product Walkthrough
If you are evaluating PALM for a new deployment, email support@spadexinc.com to schedule a guided architecture walkthrough with the SpadEx team.
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.