1. Enable GenAI

GenAI capabilities must be included in the organization’s elDoc subscription and configuration. Depending on the deployment model, GenAI can be configured with cloud-based, private, or locally deployed LLMs. elDoc is LLM-agnostic, and multiple LLMs can be configured within the same environment for different AI use cases.

2. Configure AI Models

Administrators first configure the LLM providers and models available within the elDoc environment. This defines the general pool of AI models that can be used across the platform.

To add or manage LLM providers, click the gear icon in the top-right corner, navigate to Integrations, and select Add LLM Provider. Multiple LLM providers and models can be configured and made available simultaneously. Depending on the selected Cloud, Hybrid, or On-Premise deployment model, different LLM providers and models may be available for configuration. For assistance with selecting or configuring LLMs for your deployment, please contact the elDoc Support Team at support[@]eldoc.ai.

Once configured, LLMs can be used in different ways:

  • Main Interface – AI Chat: Users can select from the LLM models made available by the administrator and use them for AI-powered retrieval, search, analysis, and interaction with enterprise knowledge.

  • AI File Management – AI Chat & AI Agents: Users can select from the available LLM models to interact with documents and instruct AI Agents to perform document-related tasks, depending on the platform configuration.

  • AI Document Processing - AI Document Understanding & Processing: Administrators can configure which LLM should be used for each specific document-processing use case. This is configured in the AI Document Processing Module under Recognition Rules → AI Data Extraction Model. For example, KYC processing can use one LLM, invoice processing another, and claims processing a different model, allowing the AI model to be selected according to the requirements of each document type or processing scenario.

This makes the distinction clearer: administrators define which LLMs are available globally, users select from those models for interactive AI capabilities, and administrators can additionally assign specific LLMs to individual GenAI Processing use cases.

This allows multiple LLMs to operate in parallel within the same elDoc environment, while administrators maintain control over which models are available and how they are used.

The selected LLM can affect processing speed, reasoning capabilities, performance, and accuracy, allowing the most appropriate model to be used for each AI task or document-processing scenario.

For more detailed information and configuration guidance, please refer to: elDoc Documentation