Fastino Releases GLiNER2.5: A Boundary-Prediction Architecture That Removes Span Enumeration From Information Extraction
Information extraction teams face a recurring choice. Small encoder models are cheap but rigid, and large language models are flexible but expensive per document. Fastino released GLiNER2.5 to narrow that gap. The release replaces span enumeration with boundary prediction: the model scores where an entity starts and ends instead of scoring every candidate span against a width grid. That single change removes the maximum entity width, allows a 4,096-word context, and keeps computation linear in
Get smart on it
Information extraction is the task of automatically finding and labeling specific pieces of information in text. A new model replaces the previous method of checking every possible text span against a schema with a boundary-prediction approach that scores where entities start and end instead, which removes limits on entity length and allows processing of much longer documents while keeping computation efficient. Three model checkpoints ranging from 74M to 287M parameters are available under an open license and can run on standard computers without specialized hardware. The approach also enables new capabilities like joint entity and relation extraction with guaranteed valid outputs, constrained classification across multiple tasks, and per-span attributes, while achieving comparable or slightly improved performance on benchmark tests compared to the previous version.
Related news
IBM's new Granite 4.2 models ride the wave of interest in local LLMs
IBM has rolled out the newest models in its family of open-weight large language models designed to be downloaded and self-hosted. The newly launched Granite 4.2 comes in 3B, 8B, and 30B parameter variants. Like previous versions, IBM is taking a decoder-only approach here. These new releases offer a 128,000-token context window natively. The 8B and 30B variants (not the 3B one) also go through an agentic reinforcement learning block; they were trained for expanded capabilities like using the t
IBM Releases Granite 4.2: Bringing Native Reasoning and Agentic RL to Open Enterprise Models
IBM has released Granite 4.2, a family of open reasoning language models in 3B, 8B, and 30B parameter sizes. Unlike earlier Granite releases, which were instruction-following assistants, Granite 4.2 is built around explicit reasoning. Every model can emit a chain of thought before answering, and every model exposes a thinking / non-thinking switch plus a low-effort mode that spends a short reasoning budget on easy questions. The models are decoder-only dense transformers, pre-trained from scrat
