GLiNER
GLiNER is a span-based named entity recognition model that supports zero-shot inference, enabling recognition of arbitrary entity types without domain-specific fine-tuning.
Timeline (last 90 days)
GLiNER 是一类面向通用命名实体识别的轻量模型，使用双向 Transformer 编码器同时处理实体类型与文本，无需为每种实体单独训练。
创作者最初计划使用 GLiNER 决策模型，但发现可本地运行的 GLiNER 并非开源，于是转向 GLiNER 2.5。
Agentic GraphRAG's entity extraction is driven by a user-defined schema, with local extraction using GLiNER and structured LLM extraction handled by BAML on selected chunks.
GLiNER is a zero-shot named entity recognition model based on a bidirectional encoder that accepts an arbitrary list of entity types at inference time without fine-tuning a model for each domain.
Entity extraction in Agentic GraphRAG can use a locally running GLiNER model, or BAML can be used for structured LLM extraction on selected chunks.
On the Deathmatch map, GLiNER with an LLM coach more than doubled its score; local GLiNER achieves approximately 8.8 decisions per second, whereas the hosted Jev is limited to only 2.5 decisions per second due to network latency.
GLiNER is a family of models whose name derives from named entity recognition, with the earliest version tracing back to 2023.
GLiNER does not provide calibrated probabilities, does not give a full probability distribution, and requires switching between different models for different use cases (GLiNER 2.5, GlyGuard, etc.).
Bidirectional encoder models can simultaneously see left and right context at every position, making them suitable for understanding tasks such as classification and entity recognition; both GLiNER and Jev are based on the bidirectional encoder approach.
The zero-shot span-based GLiNER model almost completely fails on entity types that depend on institutional function, achieving a micro F1 of only 0.243.
1 more timeline events
All Facts (11)
GLiNER 是一类面向通用命名实体识别的轻量模型，使用双向 Transformer 编码器同时处理实体类型与文本，无需为每种实体单独训练。
50%Unverified创作者最初计划使用 GLiNER 决策模型，但发现可本地运行的 GLiNER 并非开源，于是转向 GLiNER 2.5。
50%UnverifiedAgentic GraphRAG's entity extraction is driven by a user-defined schema, with local extraction using GLiNER and structured LLM extraction handled by BAML on selected chunks.
50%UnverifiedGLiNER is a zero-shot named entity recognition model based on a bidirectional encoder that accepts an arbitrary list of entity types at inference time without fine-tuning a model for each domain.
50%UnverifiedEntity extraction in Agentic GraphRAG can use a locally running GLiNER model, or BAML can be used for structured LLM extraction on selected chunks.
50%UnverifiedOn the Deathmatch map, GLiNER with an LLM coach more than doubled its score; local GLiNER achieves approximately 8.8 decisions per second, whereas the hosted Jev is limited to only 2.5 decisions per second due to network latency.
50%UnverifiedGLiNER is a family of models whose name derives from named entity recognition, with the earliest version tracing back to 2023.
50%UnverifiedGLiNER does not provide calibrated probabilities, does not give a full probability distribution, and requires switching between different models for different use cases (GLiNER 2.5, GlyGuard, etc.).
50%UnverifiedBidirectional encoder models can simultaneously see left and right context at every position, making them suitable for understanding tasks such as classification and entity recognition; both GLiNER and Jev are based on the bidirectional encoder approach.
50%UnverifiedThe zero-shot span-based GLiNER model almost completely fails on entity types that depend on institutional function, achieving a micro F1 of only 0.243.
50%UnverifiedGLiNER is a class of open NER models that supports zero-shot recognition of entity types through natural language descriptions, requiring no retraining for each new type.
50%