Fastino Labs Released GLiNER2.5-Decide AI Model
The new non-generative classifier is designed for operational decision-making tasks like triage and tool selection.
Updated on Sept. 25, 2026 in Artificial Intelligence
Live Poll
Do you believe specialized small AI models are more practical for business tasks than large models?
Fastino Labs has launched GLiNER2.5-Decide, a specialized artificial intelligence model built on a DeBERTa-v3-large encoder. The system is engineered to handle precise operational tasks such as routing and system guardrails.
Why it matters
By utilizing joint decoding to enforce rules across answers, the model provides a structured alternative to standard generative AI for complex decision-making workflows. Its ability to run in air-gapped environments offers increased security for sensitive enterprise operations.
The primary GLiNER2.5-Decide model features 340 million parameters and performs at a p50 latency of 38.3 ms on an NVIDIA V100 GPU. It is joined by a larger 1B parameter version and a 287M parameter multilingual iteration.
The players
Fastino Labs
Fastino Labs is a technology company specializing in the development of artificial intelligence models and machine learning frameworks.
The details
The model functions as a non-generative classifier that scores permitted answers before using a constrained decoder to select the highest-scoring joint assignment. It operates under an Apache 2.0 license and supports local fine-tuning through the GLiNER API.
Timeline
Fastino Labs released the GLiNER2.5-Decide model in September 2026.
The Tech Race
This release follows a trend of developers moving away from bulky generative models in favor of compact, highly efficient classifiers for mission-critical tasks. It positions Fastino Labs against providers of heavier, cloud-dependent AI systems by offering a model capable of running locally or in air-gapped environments.
For developers, this model offers a way to implement fast and accurate triage and routing systems without the latency or privacy concerns of cloud-based APIs. Users may experience more reliable tool selection and automated guardrails in enterprise applications powered by this technology.
The takeaway
The move toward specialized, smaller models suggests a growing industry preference for reliability and operational control over pure scale. Organizations can now deploy complex decision-making tools locally, reducing reliance on internet connectivity and third-party data processing.
Further reading
Learn more about advancements in Artificial Intelligence as developers prioritize efficiency and accuracy in machine learning models.
Source note: This article includes information reported by MarkTechPost.
Live Poll
Do you believe specialized small AI models are more practical for business tasks than large models?
