Sold by: Consilient
Deployed on AWS
AI products for financial crime detection, investigation and intelligence, built on patterns learned across participating institutions. They run in your environment and share only model updates, never customer or transaction data. Use them on their own or connect them for stronger detection.
Overview
Consilient provides a suite of AI capabilities for financial crime detection, investigation, and intelligence. Institutions can deploy each capability independently or connect them to strengthen how financial crime risk is discovered, modeled, assessed, and acted upon. At the core of Consilient is a privacy-preserving federated approach that enables financial institutions to learn from wider patterns of financial crime without sharing, moving, or aggregating customer or transaction data. Models run locally inside each participating institution, while approved model learning can be combined across the network and returned as stronger detection capability. Each institution retains control of its data, governance, thresholds, and compliance decisions. Federated Learning Models provide explainable machine-learning capability across core AML/CFT, correspondent banking, fraud, high-risk jurisdictions, and a growing range of financial crime behaviors. Models can operate alongside existing transaction-monitoring and case-management systems as a risk-scoring and prioritization layer, helping teams identify anomalous behavior, prioritize higher-risk activity, expand coverage, and focus investigation effort where it matters most. SAR Analyzer uses LLM-assisted discovery, natural-language search, and similarity matching across SAR/STR narratives to identify patterns, explore emerging risks, and turn prior reporting into reusable prompts, model features, rules, and detection intelligence. Synthetic Data creates realistic, risk-specific financial crime data for model training, augmentation, validation, and testing, helping address scarce or emerging behaviors where sufficient real-world examples may not yet exist. It can also be strengthened through federated learning and wider institutional experience. OSINT Agent combines entity resolution, open-source intelligence, adverse media, sanctions, company information, and model output to enrich risk assessment and support faster, more consistent review. Together, the capabilities create a reinforcing approach to financial crime detection. Institutions can start with the capability they need, integrate with existing controls, and expand over time as requirements evolve.
Highlights
- Shared learning without shared data. Models train locally inside each participating institution, while only approved model learning is combined across the federated network. Institutions retain control of customer data, governance, and compliance decisions
- Broader financial crime detection. A library of explainable models supports core Broader financial crime detection. A library of explainable models supports core AML/CFT, correspondent banking, fraud, high-risk jurisdictions, and rare or emerging financial crime behaviors, helping teams improve prioritization, expand risk coverage, and focus investigative effort.
- Stronger detection. Synthetic Data provides scarce training examples, SAR Analyzer discovers patterns in protected narratives, and OSINT Agent adds external context to risk assessment and model output.
Details
Sold by
Delivery method
Supported services
Delivery option
Consilient pipeline container (ECS Fargate, EKS, Docker)
Latest version
Operating system
Linux
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 the duration and terms of your contract with the vendor. This entitles you to a specified quantity of use for the contract duration. If you choose not to renew or replace your contract before it ends, access to these entitlements will expire.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Description | Cost/12 months |
|---|---|---|
Federated Learning Models | A suite of AML and fraud detection models that arrive already trained on patterns learned across participating institutions. They detect specific typologies, such as Structuring, Funneling and Human Trafficking, that no single institution can model from its own data. Models keep learning as the network learns, so detection keeps pace with emerging risks and regulatory expectations. The software runs in your environment and trains on your data locally. Only model updates are shared, never customer or transaction data. | $500,000.00 |
SAR Analyzer | AI-assisted analysis of SAR and STR narratives, informed by typology patterns learned across participating institutions. Search narratives in natural language, find similar and related cases, and surface patterns you would not see in your own filings alone. Turn what you find into prompts, features, rules and models, and test them consistently. Narratives stay in your environment. Only typology patterns are shared, never the narratives themselves. | $250,000.00 |
Synthetic Data | Synthetic financial crime data for training, augmenting and testing detection models where labelled data is scarce or unavailable. Scenarios are built from typology patterns learned across participating institutions and evolve as new patterns are discovered, so your models can be trained and tested against emerging risks. Customer data stays protected. | $250,000.00 |
OSINT Agent | An AI agent that combines the outputs of detection models trained on patterns learned across participating institutions with external risk information, including adverse media, sanctions and company data. It produces a consistent risk assessment to enrich alerts, speed up investigations, and support model development and monitoring. | $500,000.00 |
Dimensions summary
This contract listing offers four independent options billed by units. You choose each one separately based on what your institution needs. Federated Learning Models deliver pre-trained AML and fraud detection models. SAR Analyzer adds AI-assisted review of SAR and STR narratives. Synthetic Data provides training and testing data where labeled cases are scarce. OSINT Agent combines detection outputs with external risk information. These are not tiers or sizes of one product. Each addresses a distinct stage of detection, training, investigation, or enrichment. Pricing scales with the number of units you select per option.
Top-of-mind questions for buyers
Each option bills in units, but the marketplace listing does not define what a single unit maps to for Federated Learning Models, SAR Analyzer, Synthetic Data, or OSINT Agent. Unit scope likely varies by option and deployment size. Contact the vendor to confirm how units are counted for your institution.
The four options are priced independently, not as tiers of one product. You select units separately for each option you need. Charges add together on the same invoice. The number of units you choose per option drives that option's cost. You can buy only the options your institution requires.
No. The software runs in your environment and trains on your data locally. Only model updates or typology patterns are shared across participating institutions, never customer data, transaction data, or SAR and STR narratives. Federated learning combines permitted model updates while underlying data stays within your infrastructure.
consilient.com+2
Helpful?
Vendor refund policy
All fees are non-refundable. Subscriptions are billed through AWS Marketplace for the full subscription term, and fees already paid are not refunded if you cancel or stop using the software before the term ends. Refunds are provided only where required by law or where the End User License Agreement expressly provides for one.
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
Consilient pipeline container (ECS Fargate, EKS, Docker)
Supported services: Learn more
- Amazon ECS
- Amazon EKS
Container image
Containers are lightweight, portable execution environments that wrap server application software in a filesystem that includes everything it needs to run. Container applications run on supported container runtimes and orchestration services, such as Amazon Elastic Container Service (Amazon ECS) or Amazon Elastic Kubernetes Service (Amazon EKS). Both eliminate the need for you to install and operate your own container orchestration software by managing and scheduling containers on a scalable cluster of virtual machines.
Version release notes
Consilient pipeline v3.2.1 (build ga8d8660), AWS Marketplace image revision mp1.
Pipeline: validate, train, predict and score run types; single-file model packages (model_.consilient plus the typologies.txt selector, package format 1); overlay of customer-trained models; ensemble training; PDF output report with typology summaries. v3.2.1 removes 138 legacy flat columns from the transaction schema; there is no pipeline behaviour change versus v3.2.0.
Marketplace image (mp1): identical pipeline binary; adds the AWS License Manager entitlement check at start-up, an S3 staging mode for ECS/EKS (DATA_BUCKET), a non-root user (1001), AWS CLI v2, updated base OS packages, and disables the anonymous usage telemetry of a bundled library. Known issue: the --version flag prints 2.0.0 (fixed in the next release). Model packages are delivered by Consilient outside AWS Marketplace.
Additional details
Usage instructions
PREREQUISITES. An active subscription in the AWS account that runs the software; AWS License Manager set up once in that account (console: Start using License Manager, or: aws iam create-service-linked-role --aws-service-name license-manager.amazonaws.com); an IAM role for the container (ECS task role, EKS service-account role or EC2 instance profile) allowing license-manager:CheckoutLicense, CheckInLicense, ExtendLicenseConsumption, GetLicense and ListReceivedLicenses; Docker or ECS/EKS on x86_64; 16 GB RAM for inputs under 2 GB (32 GB for 2-5 GB, 64 GB for 10-20 GB), 100 GB disk; your Consilient model package (model_<version>.consilient, typologies.txt), delivered by Consilient with the QA guide (support@consilient.com ).
-
PULL THE IMAGE aws ecr get-login-password --region us-east-1 | docker login --username AWS --password-stdin 709825985650.dkr.ecr.us-east-1.amazonaws.com docker pull 709825985650.dkr.ecr.us-east-1.amazonaws.com/consilient/cr:v3.2.1-ga8d8660-mp1
-
LICENSE CHECK. At start-up the container calls AWS License Manager CheckoutLicense in your account to verify the subscription (dimension FederatedLearningModels) and returns the entitlement at exit. No customer data leaves your environment; the pipeline makes no other network calls. On failure the container exits with code 78 and names the missing permission or entitlement.
-
RUN WITH DOCKER (mount mode). Mount a config file, the input data, the model package directory and an output directory, and AWS credentials for the license check (mounted ~/.aws as shown, AWS_* variables or an EC2 instance profile). The image runs as user 1001: use --user "$(id -u):$(id -g)" or make the output directory writable by that user. docker run --rm --user "$(id -u):$(id -g)" -v "$HOME/.aws:/tmp/.aws:ro" -e AWS_PROFILE=default -e AWS_REGION=us-east-1 -v "$PWD/config.yaml:/config.yaml:ro" -v "$PWD/input_data:/data/input:ro" -v "$PWD/models:/data/models:ro" -v "$PWD/output:/data/output" 709825985650.dkr.ecr.us-east-1.amazonaws.com/consilient/cr:v3.2.1-ga8d8660-mp1 --config-file /config.yaml --run-type validate --validate-target input --target-variable case --run-id validate_input Run types: validate, train, predict, score (predict also writes the PDF report when run_red_flags and confidence_threshold are set). See the QA guide for each run type.
-
RUN ON ECS/FARGATE OR EKS (S3 mode). Set DATA_BUCKET (prefix overrides: see the README) plus RUN_TYPE, RUN_ID and EXTRA_ARGS: the container stages input_data/, models/, config/ and aux/ from the bucket into /data, runs, and syncs /data/output back to OUTPUT_PREFIX. Task role: read on input_data/, config/, aux/; read-write-delete on models/, output/; the License Manager actions. Fargate max 16 vCPU / 120 GB; set stopTimeout 120 and initProcessEnabled true.
-
ONE-CLICK CLOUDFORMATION. The linked template creates a VPC (public, or private with VPC endpoints and no internet route), an S3 data bucket, an ECR copy of this image in your Region (made by CodeBuild with your subscription credentials), an ECS Fargate task definition and two Step Functions state machines running validate, train, predict, validate, score and report (or the full QA chain). Upload data and model package to the bucket prefixes in the stack outputs, start the state machine, read results from s3://<bucket>/output/<execution>/. IAM roles: ECS task execution, ECS task (bucket prefixes, License Manager), CodeBuild (ECR pull/push), two Lambda functions, Step Functions. Encryption: S3 SSE-S3, ECR AES-256, CloudWatch Logs default; no keys created. Costs beyond the subscription: Fargate vCPU/GB-hours, ephemeral storage above 20 GB, CodeBuild minutes, ECR/S3 storage, CloudWatch Logs, Step Functions and, in private mode, VPC endpoints (about USD 58/mo).
-
LOGS AND EXIT CODES. The container logs to stdout (CloudWatch Logs via awslogs). 0 success, 1 pipeline or validation failure, 2 usage error, 78 license check failed, 137 out of memory. Support: support@consilient.com .
Resources
Vendor resources
Support
Vendor support
Consilient provides support by email at support@consilient.com , and through our support portal at https://consilient.com/support . Email support is available 24 hours a day, 7 days a week. Incidents are classified by severity, with response and resolution times as follows: Complete software failure or unusable output: response within 12 business hours, resolution within 24 business hours. Significant degradation of software performance or output quality: response within 12 business hours, resolution within 48 business hours. Minor issues with software operation or output: response within 24 business hours, resolution within 14 days. Buyers can also submit service requests for output queries, deployment or execution questions, deployment or execution changes, and material software changes. These are acknowledged within 24 business hours. Requests involving output or deployment questions are typically resolved within 7 days. Requests involving changes to deployment, execution, or software functionality are evaluated case by case, with timelines communicated in writing. To help resolve incidents quickly, buyers should be prepared to provide runtime logs, runtime configuration details, access to relevant data files, and a description of business impact. Consilient monitors and reports on its own compliance with these service levels, and issues monthly service reports covering any instances of non-compliance.
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
AI-native platform that automates financial crime compliance - from sanctions screening to complex trade finance investigations - with governed, evidence-based AI agents.
Shield is a unified, cloud-native platform that delivers comprehensive digital communications governance, archiving, and compliance capabilities. Designed with a modular architecture, Shield enables organizations to scale flexibly while meeting evolving regulatory and operational requirements. The platform supports ingestion from over 100 communication data sources and provides robust record management, retention, supervision, and compliance eDiscovery tools. Powered by advanced, multi-layered AI, Shield offers deep insights and operational efficiency across the full lifecycle of digital communications governance.
Fraudio is a next-generation, AI-native risk platform built for modern payment ecosystems. Delivering real-time intelligence powered by advanced machine learning and a centralized risk dataset, Fraudio leverages powerful network effects to detect complex fraud patterns across merchants, transactions, and payment players that siloed systems cannot see.
By combining best-in-class AI with a flexible, comprehensive real-time rule engine, Fraudio provides accurate, fully explainable decisions for fraud prevention, merchant monitoring, and regulatory compliance. Trusted by large banks, acquirers, PayFacs, and fintechs globally - processing billions of transactions - Fraudio enables secure growth and operational scalability within one unified platform.
Feedzai, founded in 2011, is the only end-to-end AI-native platform that stops fraud and financial crime. Unlike traditional solutions, Feedzai is smart enough to stay ahead of criminals and flexible enough not to hold back banking innovation.
With Feedzai, Financial Institutions can know customers better than they know themselves, by creating individual risk profiles and advanced analytics to assess millions of data points with millisecond latency. Feedzai specializes in scalable solutions for Banks and Fintechs, enabling them to detect and prevent evolving threats such as scams, synthetic identity fraud, and account takeovers.
Feedzai is trusted by leading Financial Institutions to manage critical risk and compliance processes, safeguarding trillions of dollars of transactions while improving customer trust and protecting consumers.
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.