Automate complex workflows involving documents of any format, type, or design with the Unstract edition of your choice: cloud, on-premise, or open-source.
Fully managed, scalable, and updated in real-time. You extract value from your documents while we handle everything else. Ideal for companies that prioritize speed and scale. And Enterprises that value ease of use and deployment.
Supported Volume
Core Platform
Prompt Studio
API Workflows
ETL & Task Pipeline Workflows
Multi-LLM Support
Core Data Connectors
Advanced Capabilities
SinglePass & Summarized Extraction
i
Simplifies prompts and documents to save LLM usage token costs.
Learn more
Security & Compliance
SSO
User Access Management
Compliance
Uptime SLA
Infrastructure
Deployment Environments
Support
Data Residency Options
Plans/Features | Starter | Growth | Enterprise |
---|---|---|---|
Supported Volume | 5,000 pages per month or 60,000 pages per year |
25,000 pages per month or 300,000 pages per year |
Custom |
Pricing | Learn more | Learn more | Learn more |
Core Platform | |||
Prompt Studio i | |||
API Workflows | |||
ETL & Task Pipeline Workflows |
|||
Multi-LLM Support | |||
Core Data Connectors | |||
Advanced Capabilities | |||
LLMChallenge i | |||
SinglePass & Summarized Extraction i |
|||
Human-in-the-Loop i | |||
Security & Compliance | |||
SSO | |||
User AccessManagement | Standard | Standard | Standard |
Compliance | Standard | Standard | SOC 2, HIPAA,ISO 27001,GDPR |
Uptime SLA | Standard | Standard | 99.999% |
Infrastructure | |||
Deployment Environments | Single | Single | 3 (Dev + Staging+ Production) |
Support | Slack Community Support | Slack Community Support | Priority, Dedicated Support |
Data Residency Options | US/EU | US/EU | US/EU |
Unstract’s native text extraction engine. Process up to 100 pages per day for free, or get increased limit on a pay-as-you-go basis.
You can also self-host with the LLMWhisperer On-Premise Enterprise edition.
Native Text
Low Cost
High Quality
High Quality with form elements
ADD-ON Source Document Highlighting API
Lesser Human
Touch points
Accurate
Document Extraction
Token
Cost Savings
Reduction in
Prompt Maintenance
The BYOK model separates the cost of the Unstract platform from the cost of the underlying AI services. You pay a subscription fee for our software and managed services (the Starter or Growth plan). You then connect your own accounts for LLMs (e.g., OpenAI, Anthropic), databases and text extractors. You are billed by those providers for your usage, at their standard rates, with no markup from us.
You choose: US or EU-based servers. Data is encrypted at rest (AES-256) and in transit (TLS 1.3). We never train models on your data. If you want data on your own infrastructure, Unstract On-Premise is the recommended solution.
Unstract supports a wide range of LLM providers: OpenAI, Anthropic, Google VertexAI, Azure OpenAI, and more! It also connects to popular vector databases like Qdrant, Weaviate, and Pinecone. Supported data sources are AWS S3, Google Drive, and Snowflake.
Deploy Unstract inside your own data centers or any private cloud infrastructure that supports Kubernetes. Get every enterprise feature while maintaining complete data sovereignty.
Air-gapped deployment
Complete network isolation for maximum security. Unstract officially supports AWS, Azure and GCP cloud environments.
Unlimited processing
Process any number of documents.
No volume restrictions or per-page pricing.
Full customization
White-label options and custom integrations.
Regulatory compliance
Meet the strictest industry standards.
Dedicated support
Implementation team and assistance when you need it.
Unstract’s native text extraction engine. Process up to 100 pages per day for free, or get increased limit on a pay-as-you-go basis.
You can also self-host with the LLMWhisperer On-Premise Enterprise edition.
Native Text
Low Cost
High Quality
High Quality with form elements
ADD-ON Source Document Highlighting API
Lesser Human
Touch points
Accurate
Document Extraction
Token
Cost Savings
Reduction in
Prompt Maintenance
Not complex at all. If you have Kubernetes running, deployment takes only a while with our deployment scripts and documentation. We support AWS, Azure, and GCP environments, plus any private cloud that runs Kubernetes. You get dedicated support during setup (and after too) to ensure everything runs smoothly from day one.
You control the update cycle completely. We provide containerized updates monthly, but you decide when to deploy them based on your change management processes. Updates are tested across 3 environments before going live. Critical security patches are flagged separately for priority deployment
You’ll need a Kubernetes cluster with minimum 16 vCPUs and 64GB RAM for production workloads. The beauty of On-Premise? No per-page pricing or volume restrictions—process millions of documents for the same infrastructure cost.
Absolutely. Unstract On-Premise supports complete network isolation—no internet connectivity required after initial deployment. All LLM processing happens within your infrastructure using self-hosted models.
The complete Unstract engine, available under the AGPL 3.0 license. Ideal for developers, research institutions, and organizations that prioritize transparency and control. Build, modify, and deploy without restrictions.
What’s included
What’s not included
Unstract’s native text extraction engine. Process up to 100 pages per day for free, or get increased limit on a pay-as-you-go basis.
You can also self-host with the LLMWhisperer On-Premise Enterprise edition.
Native Text
Low Cost
High Quality
High Quality with form elements
ADD-ON Source Document Highlighting API
Lesser Human
Touch points
Accurate
Document Extraction
Token
Cost Savings
Reduction in
Prompt Maintenance
No catch—it’s the complete Unstract engine under AGPL 3.0 license. You’re missing managed services: no LLMChallenge for complex extractions, no Human-in-the-Loop workflows, no SSO, and limited to single-user access. But the core extraction engine, API workflows, and multi-LLM support? All there. Perfect if you have engineering resources to self-manage.
If your team can deploy and manage Docker containers, you’re good to go. Basic requirements: familiarity with Python, REST APIs, and container orchestration. The 5.7K+ GitHub stars community is incredibly active—most setup questions are answered within hours.
Yes, with standard AGPL 3.0 obligations. You can use Unstract internally for any commercial purpose. If you modify the code and offer it as a service to external users, you must open-source your modifications. Using it as-is for internal document processing? No restrictions
LLMWhisperer is open-source too! It can be self-hosted and used along with Unstract for complete control. You can also choose the Cloud edition which follows a pay-as-you-go model or opt for the On-Premise Enterprise edition.
Privacy policy | Terms of service | DPA | 2025 Zipstack.Inc. All rights reserved.
We use cookies to enhance your browsing experience. By clicking "Accept", you consent to our use of cookies. Read More.