COVINA, Calif., Nov. 05, 2026 (GLOBE NEWSWIRE) — Ragic, a no-code database builder with built-in AI that enables businesses to create custom business systems and automate workflows without any programming knowledge, today announced that Ragic AI Agents now automatically use Jev, a specialized decision model from TypeSafe AI, instead of a general-purpose LLM, for qualifying classification tasks. The new integration helps businesses reduce the cost of repetitive AI automation without requiring users to select models, write code, or configure additional integrations.
As businesses increasingly adopt AI agents, even simple tasks such as categorizing customer inquiries, assigning support ticket priorities, or flagging potential policy violations can generate unexpected costs. These tasks often require little more than selecting an answer from a predefined list. However, when an AI Agent runs every time a database record is created or updated, thousands of individual AI calls can quickly add up.
Ragic’s new integration addresses this challenge by automatically identifying AI Agents that can use Jev instead of a general-purpose large language model (LLM). Jev is designed to make decisions from predefined options rather than generate free-form text, making it a more cost-effective alternative for certain classification tasks. When users save an AI Agent, Ragic checks whether it can use Jev and automatically selects the model when appropriate. Users are notified of the selection directly in the interface.
Reducing the Hidden Costs of AI Automation
Ragic’s own experience with AI Agents shows how certain AI tasks can cost far more than expected. The company’s demo-follow-up Agent helps its sales team track customer inquiries, update categories, and generate follow-up tasks. Although the team initially expected the Agent to cost less than $1 per month, it ran far more often than expected, more than 10,000 times over several weeks. At approximately $0.02 per run using Gemini Flash, the total cost reached about $200.
Based on model pricing, Ragic estimates that switching this type of classification task to Jev could reduce model costs by approximately 97–98%.
“AI costs can add up quickly, even when agents are performing relatively simple tasks,” said Jeff Kuo, CEO and Founder of Ragic. “An agent that only needs to choose between a few options doesn’t always need a large language model. By automatically selecting Jev when appropriate, we help businesses reduce unnecessary AI costs without asking users to figure out when Jev is the right fit or change how they work”
These internal comparisons demonstrate the potential savings from using specialized decision models for repetitive classification work. Actual savings will depend on the task, model usage, and pricing.
Making Cost-Efficient AI Accessible Without Programming
While specialized decision models can reduce the cost of routine AI tasks, implementing them can require technical knowledge. Businesses typically need to identify which tasks can use a decision model, connect it through an API or automation platform, and determine when a general-purpose LLM is still necessary.
Ragic eliminates these additional steps by handling model selection automatically within its platform.
The new integration offers several benefits:
- Automatic model selection – The Ragic system evaluates AI Agents when they are saved and automatically selects Jev for qualifying classification tasks. Users do not need to know whether an AI Agent should use Jev.
- Lower AI operating costs – By using Jev for simple decisions rather than relying on general-purpose LLMs for every task, businesses can reduce the cost of high-volume classification workflows.
- Native database integration – Ragic AI Agents operate within the existing Ragic database environment, allowing users to maintain their established data structures, field permissions, and business processes without additional integrations.
- No coding or technical configuration required – Ragic decides behind the scenes when an AI Agent can use Jev, allowing non-technical employees to configure and use AI Agents without writing code or designing complex AI workflows.
“We’ve always believed that the people who understand a business process should be able to build and manage the systems needed without becoming developers,” Kuo stated. “The same principle applies to AI. Ragic handles that complexity so they can focus on their business.”
Ragic AI Agent works natively inside a user’s database, where it can read records, complete forms, run workflows, and apply business logic. With automatic Jev selection, Ragic extends these capabilities by helping businesses apply a more cost-effective model to routine classification tasks while continuing to support general-purpose LLMs for more complex work.
For more information about Ragic AI Agent, visit https://www.ragic.com/intl/en/product-ai-agent?campaign=jev-pr-202610v2
About Ragic
Ragic is a no-code database builder with built-in AI that allows its users to build their own system according to their own workflow. Users can get a head start with AI, then refine their system in a quick and intuitive spreadsheet-like interface, building anything from small contact management systems to fully fledged ERP systems. Any of these systems can work with Ragic AI Agents to automate even more of the work.
Contact:
Michele Nachum
Senior Media Specialist
Firecracker PR
michele@firecrackerpr.com
Media gallery
