{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"hf-model-xenova--bart-large-mnli","slug":"xenova--bart-large-mnli","name":"bart-large-mnli","type":"model","url":"https://huggingface.co/Xenova/bart-large-mnli","page_url":"https://unfragile.ai/xenova--bart-large-mnli","categories":["model-training"],"tags":["transformers.js","onnx","bart","text-classification","zero-shot-classification","base_model:facebook/bart-large-mnli","base_model:quantized:facebook/bart-large-mnli","region:us"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"hf-model-xenova--bart-large-mnli__cap_0","uri":"capability://data.processing.analysis.zero.shot.text.classification.with.natural.language.premises","name":"zero-shot text classification with natural language premises","description":"Classifies text into arbitrary user-defined categories without task-specific fine-tuning by reformulating classification as an entailment problem. Uses BART's sequence-to-sequence architecture trained on MNLI (Multi-Genre Natural Language Inference) to compute entailment scores between input text and candidate labels, enabling dynamic category assignment at inference time without retraining.","intents":["classify customer feedback into sentiment categories without labeled training data","automatically tag documents with domain-specific labels that weren't in the original training set","detect intent from user queries against a dynamic set of possible intents defined at runtime","perform multi-label classification where the set of possible labels changes per inference call"],"best_for":["teams building rapid prototypes that need classification without labeled datasets","applications with evolving label schemas that can't afford retraining cycles","low-resource domains where gathering labeled training data is prohibitively expensive","developers integrating classification into browser-based or edge applications via ONNX"],"limitations":["inference latency is 3-5x higher than task-specific fine-tuned classifiers due to entailment computation per label","accuracy degrades with vague or ambiguous label names — requires well-crafted, semantically distinct premises","no support for hierarchical or multi-level classification without manual premise engineering","ONNX quantization reduces model size but may impact accuracy on edge cases by 1-3% depending on quantization level","batch processing is limited by ONNX.js runtime memory constraints in browser environments"],"requires":["transformers.js library (for browser/Node.js inference) or transformers Python library (for server-side)","ONNX Runtime or compatible ONNX inference engine","minimum 2GB RAM for full model, 512MB for quantized ONNX variant","Node.js 14+ or modern browser with WebAssembly support for transformers.js"],"input_types":["plain text (single or batch)","text with optional hypothesis/premise templates for custom entailment framing"],"output_types":["classification scores (logits) per candidate label","ranked label predictions with confidence scores","entailment/contradiction/neutral probability distributions"],"categories":["data-processing-analysis","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hf-model-xenova--bart-large-mnli__cap_1","uri":"capability://data.processing.analysis.onnx.quantized.model.inference.for.edge.and.browser.deployment","name":"onnx-quantized model inference for edge and browser deployment","description":"Provides a quantized ONNX (Open Neural Network Exchange) version of BART-large-mnli that reduces model size from ~1.6GB to ~400-500MB while maintaining inference capability on CPU-only devices and browsers. Uses 8-bit or mixed-precision quantization to compress weights and activations, enabling deployment in resource-constrained environments without GPU acceleration.","intents":["deploy text classification directly in the browser without backend API calls","run inference on edge devices (mobile, IoT, embedded systems) with limited memory","reduce model serving costs by eliminating cloud inference infrastructure for classification tasks","build offline-capable applications where classification must work without internet connectivity"],"best_for":["frontend developers building client-side NLP features with transformers.js","mobile app developers targeting devices with <1GB available RAM","teams building privacy-first applications where text cannot leave the device","edge computing deployments (Raspberry Pi, NVIDIA Jetson, industrial IoT)"],"limitations":["quantization introduces 1-3% accuracy loss on average, with higher variance on edge-case inputs","ONNX.js runtime in browsers has limited operator support — some model variants may not convert cleanly","first inference pass includes model loading overhead (200-800ms depending on network/storage), subsequent calls are faster","batch inference in browser is constrained by available RAM; typical batch size limited to 4-8 sequences","no GPU acceleration in browser environment — inference speed is 5-10x slower than GPU-accelerated server inference"],"requires":["ONNX Runtime (onnxruntime-web for browser, onnxruntime for Node.js/Python)","transformers.js library for browser integration","minimum 512MB RAM available for model loading and inference","modern browser with WebAssembly support (Chrome 74+, Firefox 79+, Safari 14.1+) or Node.js 14+"],"input_types":["text sequences (tokenized or raw)","batch inputs as arrays of text strings"],"output_types":["ONNX tensor outputs (logits, attention weights)","structured classification results with scores"],"categories":["data-processing-analysis","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hf-model-xenova--bart-large-mnli__cap_2","uri":"capability://data.processing.analysis.multi.label.entailment.scoring.with.candidate.ranking","name":"multi-label entailment scoring with candidate ranking","description":"Computes entailment scores between input text and multiple candidate labels simultaneously, ranking candidates by their entailment probability. The model processes each (text, label) pair through BART's encoder-decoder, generating logits for entailment/neutral/contradiction classes, then ranks labels by entailment confidence to support both single-label and multi-label classification scenarios.","intents":["assign multiple applicable labels to a single document (e.g., a news article tagged with both 'politics' and 'economy')","rank candidate labels by relevance confidence rather than binary accept/reject decisions","implement threshold-based multi-label classification where only high-confidence labels are assigned","debug classification decisions by inspecting entailment scores for each candidate label"],"best_for":["content moderation systems requiring multiple violation categories per item","document tagging systems where items naturally belong to multiple categories","recommendation systems that need to score items against many candidate attributes","developers building interpretable classification pipelines where score transparency is required"],"limitations":["computational cost scales linearly with number of candidate labels — 100 labels = ~100x inference cost vs single label","no built-in handling of label dependencies or hierarchies — labels are scored independently","threshold selection for multi-label assignment is application-specific and requires tuning","entailment scores are not directly comparable across different label sets due to BART's softmax normalization per input"],"requires":["transformers.js or transformers library with BART model support","ability to construct natural language premises from candidate labels","application logic to handle threshold-based label filtering and ranking"],"input_types":["text sequence (single document or query)","list of candidate labels (strings)"],"output_types":["ranked list of (label, entailment_score) tuples","entailment probability distributions per label","filtered multi-label assignments based on configurable thresholds"],"categories":["data-processing-analysis","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hf-model-xenova--bart-large-mnli__cap_3","uri":"capability://data.processing.analysis.cross.lingual.zero.shot.classification.via.transfer.learning","name":"cross-lingual zero-shot classification via transfer learning","description":"Applies zero-shot classification to non-English text by leveraging BART's multilingual pretraining and MNLI's English entailment knowledge, enabling classification in 50+ languages without language-specific fine-tuning. The model transfers entailment reasoning from English to other languages through shared token embeddings and cross-lingual attention mechanisms learned during pretraining.","intents":["classify customer support tickets in multiple languages without separate models per language","detect intent in chatbots supporting global users without language-specific training data","tag user-generated content in non-English languages without maintaining separate label vocabularies","build multilingual content moderation systems that apply consistent classification logic across languages"],"best_for":["global SaaS platforms serving users in multiple languages","teams without resources to collect language-specific labeled datasets","applications requiring consistent classification logic across language boundaries","developers building inclusive products that don't prioritize English-language performance"],"limitations":["cross-lingual transfer performance degrades for low-resource languages (e.g., Swahili, Tagalog) — expect 5-15% accuracy drop vs English","label names must be provided in English or translated to English for optimal performance; non-English label names reduce accuracy","performance is best for languages with similar linguistic structure to English; morphologically complex languages (e.g., Turkish, Finnish) see larger drops","no explicit language detection — model assumes input language matches label language, requiring application-level language handling"],"requires":["transformers.js or transformers library with multilingual BART support","English-language label definitions (or manual translation of labels to English)","awareness of language-specific performance characteristics for production SLAs"],"input_types":["text in any of 50+ supported languages","English-language label definitions"],"output_types":["classification scores per label","ranked predictions with confidence scores","language-agnostic entailment probability distributions"],"categories":["data-processing-analysis","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hf-model-xenova--bart-large-mnli__cap_4","uri":"capability://data.processing.analysis.batch.inference.with.dynamic.label.sets","name":"batch inference with dynamic label sets","description":"Processes multiple text inputs and multiple candidate labels in a single inference pass, computing entailment scores for all (text, label) combinations. Implements batching at both the text and label levels, optimizing throughput by reusing model computations across inputs while supporting different label sets per text input without model reloading.","intents":["classify a batch of 100+ documents with different applicable label sets in a single inference call","process streaming classification requests efficiently by batching inputs across time windows","implement efficient content moderation pipelines that classify multiple items in parallel","build recommendation systems that score items against large candidate sets (100+ labels) with minimal latency"],"best_for":["batch processing pipelines (ETL, data labeling, content ingestion)","high-throughput classification services handling 100+ requests/second","teams optimizing inference costs by amortizing model loading overhead across batches","applications with variable label sets per input (e.g., personalized recommendation)"],"limitations":["batch size is constrained by available GPU/CPU memory — typical limits are 32-64 texts × 50 labels on consumer hardware","dynamic label sets per input require careful batching logic to avoid padding inefficiencies","latency for small batches (<5 items) may be dominated by model loading overhead rather than computation","ONNX.js runtime in browsers has stricter memory limits — batch sizes typically capped at 4-8 sequences"],"requires":["transformers.js or transformers library with batch processing support","sufficient GPU/CPU memory for batch size × label count × model size","application logic to handle variable label sets and batch padding"],"input_types":["batch of text sequences (array of strings)","batch of candidate label sets (array of label lists, potentially different per input)"],"output_types":["batch of classification results (array of ranked label predictions)","batch of entailment score matrices (texts × labels)","structured batch results with per-input metadata"],"categories":["data-processing-analysis","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":36,"verified":false,"data_access_risk":"high","permissions":["transformers.js library (for browser/Node.js inference) or transformers Python library (for server-side)","ONNX Runtime or compatible ONNX inference engine","minimum 2GB RAM for full model, 512MB for quantized ONNX variant","Node.js 14+ or modern browser with WebAssembly support for transformers.js","ONNX Runtime (onnxruntime-web for browser, onnxruntime for Node.js/Python)","transformers.js library for browser integration","minimum 512MB RAM available for model loading and inference","modern browser with WebAssembly support (Chrome 74+, Firefox 79+, Safari 14.1+) or Node.js 14+","transformers.js or transformers library with BART model support","ability to construct natural language premises from candidate labels"],"failure_modes":["inference latency is 3-5x higher than task-specific fine-tuned classifiers due to entailment computation per label","accuracy degrades with vague or ambiguous label names — requires well-crafted, semantically distinct premises","no support for hierarchical or multi-level classification without manual premise engineering","ONNX quantization reduces model size but may impact accuracy on edge cases by 1-3% depending on quantization level","batch processing is limited by ONNX.js runtime memory constraints in browser environments","quantization introduces 1-3% accuracy loss on average, with higher variance on edge-case inputs","ONNX.js runtime in browsers has limited operator support — some model variants may not convert cleanly","first inference pass includes model loading overhead (200-800ms depending on network/storage), subsequent calls are faster","batch inference in browser is constrained by available RAM; typical batch size limited to 4-8 sequences","no GPU acceleration in browser environment — inference speed is 5-10x slower than GPU-accelerated server inference","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.44771447080436505,"quality":0.2,"ecosystem":0.5000000000000001,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.35,"quality":0.2,"ecosystem":0.1,"match_graph":0.3,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:22.766Z","last_scraped_at":"2026-05-03T14:22:57.756Z","last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":null,"model_downloads":62837,"model_likes":5}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=xenova--bart-large-mnli","compare_url":"https://unfragile.ai/compare?artifact=xenova--bart-large-mnli"}},"signature":"cOrCPHLOBe5qCEyhR7D1kXHHLU6FMc/SZPS76Y01sb8xl4I1QZ1PHYO8snUEQI9h8f0x96h6O7flIywtA6WzCQ==","signedAt":"2026-06-22T06:57:45.032Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/xenova--bart-large-mnli","artifact":"https://unfragile.ai/xenova--bart-large-mnli","verify":"https://unfragile.ai/api/v1/verify?slug=xenova--bart-large-mnli","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}