decision-model-preview

Use it locally to classify text for tasks such as intent detection, routing, sentiment, moderation, urgency, and document type.

Availability

Load with AutoExtractor from fastino/GLiNER2.5-Decide; install gliner2[local].

Pricing model

Not established

Deployment

Load and ship it locally.

Privacy

Not established

Licence

Not established

• It returns probability distributions and confidence scores. ^[1]

  It performs classification, yes-or-no decisions, and scoring in parallel based on text or business context, returning probability distributions and confidence scores for scenarios such as ticket routing, content moderation, agent routing, and result validation.

• It operates based on text or business context. ^[1]

  It performs classification, yes-or-no decisions, and scoring in parallel based on text or business context, returning probability distributions and confidence scores for scenarios such as ticket routing, content moderation, agent routing, and result validation.

• Its service scope is International. ^[1]

  Service scope: International

• Listed use cases include ticket routing, content moderation, agent routing, and result validation. ^[1]

  It performs classification, yes-or-no decisions, and scoring in parallel based on text or business context, returning probability distributions and confidence scores for scenarios such as ticket routing, content moderation, agent routing, and result validation.

• The 340M English model accepts any label set at call time. ^[2]

  **The 340M English classification model in the GLiNER2.5 family.** Pass any label set at call time: intent, routing, sentiment, priority, policy, and multi-label tags, in a single forward pass.

• A single forward pass can score several classification heads. ^[2]

  **The 340M English classification model in the GLiNER2.5 family.** Pass any label set at call time: intent, routing, sentiment, priority, policy, and multi-label tags, in a single forward pass.

• The model can be loaded with AutoExtractor from fastino/GLiNER2.5-Decide. ^[2]
• The model can be loaded and shipped locally. ^[2]

  **The 340M English classification model in the GLiNER2.5 family.** Pass any label set at call time: intent, routing, sentiment, priority, policy, and multi-label tags, in a single forward pass.

• The release is not a general-purpose model and does not reason, explain, or answer open questions. ^[2]

  It does not reason, explain, or answer open questions.

• The benchmark suite is English; the source recommends GLiNER2.5-multi-Decide for multilingual input. ^[2]