GLiClass: Open-Source JEV

Knowledgator released GLiClass, an open-source zero-shot sequence classification model inspired by the GLiNER framework that performs classification in a single forward pass and runs approximately 10 times faster than traditional cross-encoder models while achieving comparable performance. The model is installable via pip and supports hierarchical label structures using dot notation, in-context examples through the <<EXAMPLE>> token, custom prompts, and automatic text chunking for long documents. GLiClass is available on GitHub and Hugging Face, with a small-v1.0 model checkpoint and an interactive demo.

GLiClass is an efficient, zero-shot sequence classification model inspired by the GLiNER https://github.com/urchade/GLiNER/tree/main framework. It achieves comparable performance to traditional cross-encoder models while being significantly more computationally efficient, offering classification results approximately 10 times faster by performing classification in a single forward pass. 📄 Blog https://medium.com/@knowledgrator/pushing-zero-shot-classification-to-the-limit-696a2403032f • 📢 Discord https://discord.gg/dkyeAgs9DG • 📺 Demo https://huggingface.co/spaces/knowledgator/GLiClass SandBox • 🤗 Available models https://huggingface.co/models?sort=trending&search=gliclass • Install GLiClass easily using pip: pip install gliclass Clone and install directly from GitHub: git clone https://github.com/Knowledgator/GLiClass cd GLiClass python -m venv venv source venv/bin/activate Windows: venv\Scripts\activate pip install -r requirements.txt pip install . Verify your installation: python import gliclass print gliclass. version python from gliclass import GLiClassModel, ZeroShotClassificationPipeline from transformers import AutoTokenizer model = GLiClassModel.from pretrained "knowledgator/gliclass-small-v1.0" tokenizer = AutoTokenizer.from pretrained "knowledgator/gliclass-small-v1.0" pipeline = ZeroShotClassificationPipeline model, tokenizer, classification type='multi-label', device='cuda:0' text = "One day I will see the world " labels = "travel", "dreams", "sport", "science", "politics" results = pipeline text, labels, threshold=0.5 0 for result in results: print f"{result 'label' } = {result 'score' :.3f}" GLiClass now supports hierarchical label structures using dot notation: hierarchical labels = { "sentiment": "positive", "negative", "neutral" , "topic": "product", "service", "shipping" } text = "The product quality is amazing but delivery was slow" results = pipeline text, hierarchical labels, threshold=0.5 0 for result in results: print f"{result 'label' } = {result 'score' :.3f}" Output: sentiment.positive = 0.892 topic.product = 0.921 topic.shipping = 0.763 Get hierarchical output matching your input structure: results = pipeline text, hierarchical labels, return hierarchical=True 0 print results Output: { "sentiment": {"positive": 0.892, "negative": 0.051, "neutral": 0.124}, "topic": {"product": 0.921, "service": 0.153, "shipping": 0.763} } Improve classification accuracy with in-context examples using the <<EXAMPLE token: examples = { "text": "Love this item, great quality ", "labels": "positive", "product" }, { "text": "Customer support was unhelpful", "labels": "negative", "service" } text = "Fast delivery and the item works perfectly " labels = "positive", "negative", "product", "service", "shipping" results = pipeline text, labels, examples=examples, threshold=0.5 0 for result in results: print f"{result 'label' } = {result 'score' :.3f}" Add custom prompts to guide the classification task: text = "The battery life on this phone is incredible" labels = "positive", "negative", "neutral" results = pipeline text, labels, prompt="Classify the sentiment of this product review:", threshold=0.5 0 Use per-text prompts for batch processing: texts = "Review about electronics", "Review about clothing" prompts = "Analyze this electronics review:", "Analyze this clothing review:" results = pipeline texts, labels, prompt=prompts Process long documents with automatic text chunking: python from gliclass import ZeroShotClassificationWithChunkingPipeline chunking pipeline = ZeroShotClassificationWithChunkingPipeline model, tokenizer, text chunk size=8192, text chunk overlap=256, labels chunk size=8 long document = "..." Very long text labels = "category1", "category2", "category3" results = chunking pipeline long document, labels, threshold=0.5 With new models trained with retrieval-agumented classification, such as this model https://huggingface.co/knowledgator/gliclass-base-v2.0-rac-init you can specify examples to improve classification accuracy: example = { "text": "A new machine learning platform automates complex data workflows but faces integration issues.", "all labels": "AI", "automation", "data analysis", "usability", "integration" , "true labels": "AI", "integration", "automation" } text = "The new AI-powered tool streamlines data analysis but has limited integration capabilities." labels = "AI", "automation", "data analysis", "usability", "integration" results = pipeline text, labels, threshold=0.1, rac examples= example 0 for predict in results: print f"{predict 'label' } = {predict 'score' :.3f}" Deploy GLiClass with Ray Serve for production workloads with dynamic batching and memory-aware processing. pip install gliclass serve Default model python -m gliclass.serve Specify model and port python -m gliclass.serve --model knowledgator/gliclass-edge-v3.0 --port 8000 With config file python -m gliclass.serve --config serve configs/serve config.yaml python from gliclass.serve import GLiClassClient client = GLiClassClient url="http://localhost:8000/gliclass" result = client.classify text="This is a great product ", labels= "positive", "negative", "neutral" , threshold=0.3, print result {"label": "positive", "score": 0.95}, ... The HTTP endpoint processes one text per request. curl -X POST http://localhost:8000/gliclass \ -H "Content-Type: application/json" \ -d '{ "texts": "This is a great product ", "labels": "positive", "negative", "neutral" , "threshold": 0.3 }' Response: {"label": "positive", "score": 0.95}, ... Note: For batch processing multiple texts, use the ZeroShotClassificationPipeline directly instead of the serving API. See serve configs/serve config.yaml for full configuration options. - Sentiment Analysis: Rapidly classify texts as positive, negative, or neutral. - Document Classification: Efficiently organize and categorize large document collections. - Search Results Re-ranking: Improve relevance and precision by reranking search outputs. - News Categorization: Automatically tag and organize news articles into predefined categories. - Fact Checking: Quickly validate and categorize statements based on factual accuracy. Prepare your training data as follows: {"text": "Sample text.", "all labels": "sports", "science", "business" , "true labels": "sports" }, ... Optionally, specify confidence scores explicitly: {"text": "Sample text.", "all labels": "sports", "science" , "true labels": {"sports": 0.9}}, ... Please, refer to the train.py script to set up your training from scratch or fine-tune existing models. GLiClass supports multiple architecture types: - uni-encoder : Single encoder for both text and labels default, most efficient - bi-encoder : Separate encoders for text and labels - bi-encoder-fused : Bi-encoder with label embeddings fused into text encoding - encoder-decoder : Encoder-decoder architecture for sequence-to-sequence tasks python from gliclass import GLiClassBiEncoder Load a bi-encoder model model = GLiClassBiEncoder.from pretrained "knowledgator/gliclass-biencoder-v1.0" Configure how token embeddings are pooled: - first : First token CLS token - avg : Average pooling - max : Max pooling - last : Last token - sum : Sum pooling - rms : Root mean square pooling - abs max : Max of absolute values - abs avg : Average of absolute values python from gliclass import GLiClassModelConfig config = GLiClassModelConfig pooling strategy='avg', class token pooling='average' or 'first' Choose different scoring mechanisms for classification: - simple : Dot product fastest - weighted-dot : Weighted dot product with learned projections - mlp : Multi-layer perceptron scorer - hopfield : Hopfield network-based scorer config = GLiClassModelConfig scorer type='mlp' GLiClass supports incremental, multi-session text classification over a decoder-KV model. Instead of re-encoding the full document on every update, it maintains a persistent KV cache per session and runs the scorer only when a pluggable strategy decides classification should fire. pip install gliclass streaming python from gliclass.streaming import StreamingPipeline, SessionInput, EveryNTokensStrategy pipeline = StreamingPipeline model, tokenizer, device="cuda", max cache len=1024 strategy = EveryNTokensStrategy n=50 for chunk in text chunks: outputs = pipeline SessionInput session id="doc 001", text=chunk, labels= "science", "politics", "finance" , strategy=strategy, classification type="multi-label", if outputs 0 .triggered: print outputs 0 .predictions Built-in strategies: EveryChunkStrategy , EveryNTokensStrategy , OnDelimiterStrategy , SlidingWindowStrategy , ComposedStrategy , NeverStrategy . For full documentation on session management, KV cache internals, batching, CPU offloading, and custom strategies, see docs/streaming.md https://github.com/Knowledgator/GLiClass/blob/main/docs/streaming.md . GLiClass supports optional flash attention backends for faster inference. pip install flashdeberta DeBERTa v2 pip install turbot5 T5 / mT5 Enable via environment variable: export USE FLASHDEBERTA=1 If flashdeberta is installed, DeBERTa v2 models will use FlashDebertaV2Model . Otherwise, GLiClass falls back to DebertaV2Model . Enable via environment variable: export TURBOT5 ATTN TYPE=triton-basic If turbot5 is installed, T5 / mT5 models will use FlashT5EncoderModel . Otherwise, GLiClass falls back to T5EncoderModel . Notes: - Flash backends are optional - Enabled automatically when available - No code changes required Want it even tighter single block , or is this the sweet spot? If you find GLiClass useful in your research or project, please cite our papers: @misc{stepanov2025gliclassgeneralistlightweightmodel, title={GLiClass: Generalist Lightweight Model for Sequence Classification Tasks}, author={Ihor Stepanov and Mykhailo Shtopko and Dmytro Vodianytskyi and Oleksandr Lukashov and Alexander Yavorskyi and Mykyta Yaroshenko}, year={2025}, eprint={2508.07662}, archivePrefix={arXiv}, primaryClass={cs.LG}, url={https://arxiv.org/abs/2508.07662}, }