On October 22, Ant Digital Technologies officially released the decision model LingDT-Decision-1.0. Built specifically for fast decision-making, the model offers advantages such as higher accuracy, faster response, and lower Token costs when handling structured decision tasks, making it especially suitable for industry AI scenarios. LingDT-Decision-1.0 comes in two versions, 4B and 27B. Currently, the model is live on Ant Digital Technologies' DTMaaS platform, and developers can try it for free for a limited time.
Tests by the Ant Digital Technologies team in real business scenarios such as customer service, risk control, and finance showed that, when handling the same tasks, LingDT-Decision-1.0 reduces Token consumption by more than 90% compared with large language models, improves response time by 10x, and delivers better performance.
Compared with other decision models, LingDT-Decision-1.0 is the first to natively support multi-select and ordered decision-making capabilities, making it better suited to real enterprise application scenarios. Across 23 open-source evaluation benchmarks, LingDT-Decision-1.0's overall score exceeded popular decision models such as Microsoft-Decision-1 and Jev1.13., and the 27B version of LingDT-Decision-1.0 achieved the highest accuracy among models with comparable parameters.
Designed specifically for decision tasks, with Token consumption cut by more than 90%—faster and more accurate
In actual enterprise applications, there are large numbers of high-frequency decision-type tasks.
For example, in a customer service scenario, after receiving an inquiry such as "This order has not shipped after three days, please handle it as soon as possible," the AI needs to quickly make a series of judgments and decisions: the problem type (refund, expedite request, or complaint?), ticket urgency (does it need priority handling), owning department (which department the ticket goes to), and risk assessment (whether human takeover is needed).
Traditional large language models have clear shortcomings in these scenarios. Large language models typically need to reason token by token (Token) and then generate output text, which means a long pipeline, high cost, greater susceptibility to errors, and unstable performance.
Decision models are designed precisely to address these pain points; their goal is not to generate text but to make fast, accurate judgments. They are therefore naturally suited to high-frequency decision-type tasks in enterprise business, such as intent detection, tool selection, rule classification, and result scoring.
Currently, the LingDT-Decision-1.0 model is already being tested in real business scenarios at Ant Digital Technologies including intelligent customer service, security risk control, and financial applications. Its core idea is to shift the large model from "first generating a structured answer, then performing subsequent parsing" to "making a judgment directly among constrained options." This not only reduces formatting problems caused by unstable output, but also makes the originally cumbersome parsing and validation process much simpler.
For example, in a data annotation scenario, compared with DS-V4-flash, Token consumption drops by about 97% and processing time improves by about 10x; in an intelligent customer service scenario, compared with Qwen-3.7-flash, Token consumption drops by 90%, processing time improves by about 1x, and risk identification accuracy rises from 76% to 89%.
In financial application scenarios, for tasks such as intent recognition, tool calling, and hallucination detection, processing time improves by about 5x, full tool-combination match rates improve by about 2.6x, and hallucination recall rises from 44% to 70%, making risk detection more sensitive.
Overall, LingDT-Decision-1.0 consistently demonstrates the advantages of lower cost, faster response, and more accurate judgment across multiple real-world scenarios. At the same time, a single model can handle multiple types of judgment tasks, making integration simpler and maintenance easier, so it is easier to reuse and roll out across different business scenarios.
Building the next generation of decision models, with two features debuting globally
Compared with other decision models, Ant Digital Technologies' LingDT-Decision-1.0 places greater emphasis on being built for real enterprise application scenarios. Beyond basic capabilities such as judgment, single-select, classification, and rating, LingDT-Decision-1.0 is also the first to introduce native multi-select and ordered decision-making (DAG) features, better adapting to the complex, multi-step decision needs of enterprise business.
In enterprise scenarios, what is more common is not "picking one" but "identifying a set of items to handle and executing them in order." For example, when a user says: "I want a refund, cancel the renewal, reissue the invoice, but do not deactivate my account," single-select often only recognizes "refund" as the main request, easily missing other needs. Multi-select can identify all requests at once; ordered decision-making can further determine that the refund requires verification first, canceling the renewal can be executed first, and invoice processing may need to wait for the refund result.
Therefore, multi-select and ordered decision-making better reflect real business processes and can better improve accuracy and final processing results.
For native multi-select capability, LingDT-Decision-1.0 allows a single task to hit multiple candidates simultaneously, meeting needs such as multi-label recognition, compound intents, and tool combinations; for ordered decision-making capability, the model can complete multi-step decisions according to dependencies between tasks, closely matching real business processes in finance, risk control, customer service, and more.
Test results covering 23 mainstream open-source evaluation benchmarks show that the 27B version of LingDT-Decision-1.0 ranked first in overall score (90% x single-select score + 10% x multi-select score), with accuracy surpassing popular decision models such as Microsoft-Decision-1, Jev1.13.0, and Fastino/GLiDE.
Among them, on multi-select tasks, the 4B and 27B versions of LingDT-Decision-1.0 achieved accuracy surpassing 16 decision models including Microsoft-Decision-1Jev1.13.0 and Star-lux, demonstrating the advantage of native multi-select capability in multi-label and combination judgment tasks. On single-select tasks, it ranked in the top tier.