Apps · Python library

GLiNER

by Urchade Zaratiana, Fastino and Knowledgator · Apache-2.0 · 4,159 GitHub stars

Python libraries for GLiNER-family models: gliner for entities, gliner2 for schemas and decisions, gliclass for labels, all in one pass.

macOSWindowsLinux

Get GLiNER

At a glance

Best for
Fast label decisions and entity extraction on a CPU with labels you choose at run time.
Inference engine
  • PyTorch
  • ONNX
GPU support
  • NVIDIA CUDA
  • Apple Metal
  • CPU only
Opens
  • Safetensors (original weights)
  • ONNX
Features
  • HTTP API
Platforms
Python 3 on macOS, Windows and Linux; runs well on a CPU.
Speed
One forward pass scores every label, so a 300M model answers in milliseconds even on a CPU.

Sources: github.com · github.com · github.com

Install

macOS

pip install gliner2

Windows

pip install gliner2

Linux

pip install gliner2

6 models on this site run in GLiNER

Decision models it can load: rerankers, classifiers and scorers with a supported architecture. Open a model for its speed and memory on your hardware.

Model Made by Type Size Opens it as Quality tier
GLiNER2.5 Decide Fastino Decision 490M Original weights 80
GLiClass ModernBERT Large v3 Knowledgator Decision 400M Original weights 74
GLiGuard 300M Fastino Decision 210M Original weights 68
GLiNER2 Base v1 Fastino Decision 210M Original weights 68
GLiClass Base v3 Knowledgator Decision 190M Original weights 66
GLiNER Multi v2.1 Urchade Zaratiana Decision 290M Original weights 62

Alternatives

LM Studio Ollama llama.cpp vLLM Jan Text Embeddings Inference MLX LM SGLang Lemonade Infinity Unsloth Desktop Sentence Transformers Docker Model Runner LocalAI Atomic Chat

Which app should I use?

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.