The model catalog: what actually runs in a browser

Frequently asked questions

Why can I not use any Hugging Face model?

transformers.js runs ONNX graphs, not PyTorch. A model needs an ONNX export (usually under an "onnx/" folder in the repo) before a browser can load it. That is why the catalogue is a curated list rather than all of Hugging Face.

Which quantization should I pick?

q8 is the sweet spot for most vision and encoder models; q4 halves the download again for chat models at a small quality cost; fp32 is the safest and largest. Avoid fp16 and q4f16 unless your GPU reports the WebGPU "shader-f16" feature — many desktop GPUs on Linux do not.

Is my data uploaded to a server?

No. The model weights are downloaded from Hugging Face to your browser once, and every inference after that runs on your own GPU through WebGPU. Your text, images and audio are never sent anywhere — hjLabs.in has no inference server and no way to see your input.

Does this work on a phone?

Small models do. Recent mobile Chrome and Safari expose WebGPU, but memory is the binding limit — stay under roughly 200 MB of weights on Android, and expect iOS to reclaim the tab sooner. The catalog shows every download size up front so you can pick something that fits.

Can I use the output commercially?

That depends on the individual model, not on this site. Each model page links to its Hugging Face repository, where the licence is stated — most of the catalog is Apache-2.0 or MIT, a few are research-only. We add nothing on top: hjLabs.in claims no rights over anything you produce here.

Every model, by task (149 in total)

Image Classification

• ViT NSFW Detector — AdamCodd/vit-base-nsfw-detector, 89 MB, 86M params
• ViT Base 224 — Xenova/vit-base-patch16-224, 88 MB, 86M params
• Facial Emotion ViT — Xenova/facial_emotions_image_detection, 88 MB, 86M params
• ResNet-50 — Xenova/resnet-50, 26 MB, 25.6M params
• MobileViT Small — Xenova/mobilevit-small, 23 MB, 5.6M params
• Swin Food-101 — onnx-community/swin-finetuned-food101-ONNX, 93 MB

Image Captioning

• ViT-GPT2 Captioner — Xenova/vit-gpt2-image-captioning, 248 MB, 239M params
• DistilViT — Mozilla/distilvit, 189 MB
• ViT-GPT2 Flower Captioner — cristianglezm/ViT-GPT2-FlowerCaptioner-ONNX, 247 MB, 239M params

Zero-Shot Image Labels

• SigLIP Base 224 — Xenova/siglip-base-patch16-224, 214 MB, 203M params
• CLIP ViT-B/32 — Xenova/clip-vit-base-patch32, 157 MB, 151M params
• CLIP ViT-B/16 — Xenova/clip-vit-base-patch16, 156 MB, 150M params
• CLIP ViT-L/14 — Xenova/clip-vit-large-patch14, 436 MB, 428M params
• TinyCLIP ViT-8M — onnx-community/TinyCLIP-ViT-8M-16-Text-3M-YFCC15M-ONNX, 28 MB, 11M params
• FashionCLIP — ff13/fashion-clip, 156 MB, 151M params

Chat

• SmolLM2 135M Instruct — HuggingFaceTB/SmolLM2-135M-Instruct, 186 MB, 135M params
• SmolLM2 360M Instruct — HuggingFaceTB/SmolLM2-360M-Instruct, 392 MB, 360M params
• SmolLM2 1.7B Instruct — HuggingFaceTB/SmolLM2-1.7B-Instruct, 1416 MB, 1.7B params
• Qwen3 0.6B — onnx-community/Qwen3-0.6B-ONNX, 928 MB, 0.6B params
• Qwen2.5 0.5B Instruct — onnx-community/Qwen2.5-0.5B-Instruct, 798 MB, 0.5B params
• Gemma 3 1B Instruct — onnx-community/gemma-3-1b-it-ONNX, 880 MB, 1B params
• Qwen3 1.7B — onnx-community/Qwen3-1.7B-ONNX, 2195 MB, 1.7B params
• Llama 3.2 1B Instruct — onnx-community/Llama-3.2-1B-Instruct-ONNX, 1704 MB, 1.2B params
• Llama 3.2 3B Instruct — onnx-community/Llama-3.2-3B-Instruct-ONNX, 3417 MB, 3.2B params
• Gemma 3 270M Instruct — onnx-community/gemma-3-270m-it-ONNX, 348 MB, 270M params
• DeepSeek-R1 Distill 1.5B — onnx-community/DeepSeek-R1-Distill-Qwen-1.5B-ONNX, 1976 MB, 1.5B params
• LFM2 1.2B — onnx-community/LFM2-1.2B-ONNX, 854 MB, 1.2B params
• Qwen3 4B — onnx-community/Qwen3-4B-ONNX, 2881 MB, 4B params
• Qwen2.5 Coder 1.5B — onnx-community/Qwen2.5-Coder-1.5B-Instruct, 1932 MB, 1.5B params
• Phi-3.5 Mini (3.8B) — onnx-community/Phi-3.5-mini-instruct-onnx-web, 2320 MB, 3.8B params
• SmolLM3 3B — HuggingFaceTB/SmolLM3-3B-ONNX, 2854 MB, 3B params
• LFM2 2.6B — onnx-community/LFM2-2.6B-ONNX, 2096 MB, 2.6B params

Image Segmentation

• SegFormer B0 (ADE20K) — Xenova/segformer-b0-finetuned-ade-512-512, 15 MB, 3.8M params
• Face Parsing — jonathandinu/face-parsing, 89 MB
• SegFormer B2 (ADE20K) — Xenova/segformer-b2-finetuned-ade-512-512, 110 MB, 27M params
• DETR ResNet-50 Panoptic — Xenova/detr-resnet-50-panoptic, 172 MB, 43M params
• SegFormer B2 Clothes — Xenova/segformer_b2_clothes, 110 MB, 27M params
• SegFormer B0 Clothes — Xenova/segformer_b0_clothes, 15 MB, 3.8M params

Remove Background

• MODNet — Xenova/modnet, 26 MB, 6.5M params
• ORMBG — onnx-community/ormbg-ONNX, 176 MB
• BiRefNet Lite — onnx-community/BiRefNet_lite-ONNX, 224 MB
• BiRefNet 512 — onnx-community/BiRefNet_512x512-ONNX, 940 MB
• RMBG 1.4 — kittypdf/RMBG-1.4-transformersjs, 176 MB, 44M params

Embeddings

• mxbai-embed-large-v1 — mixedbread-ai/mxbai-embed-large-v1, 338 MB, 335M params
• all-MiniLM-L6-v2 — Xenova/all-MiniLM-L6-v2, 91 MB, 22.7M params
• BGE Base EN v1.5 — Xenova/bge-base-en-v1.5, 111 MB, 109M params
• BGE Small EN v1.5 — Xenova/bge-small-en-v1.5, 134 MB, 33M params
• Paraphrase Multilingual MiniLM — Xenova/paraphrase-multilingual-MiniLM-L12-v2, 135 MB, 118M params
• Multilingual E5 Small — Xenova/multilingual-e5-small, 135 MB, 118M params
• GTE Small — Xenova/gte-small, 134 MB, 33M params
• SapBERT — Xenova/SapBERT-from-PubMedBERT-fulltext, 111 MB, 109M params
• CODER (UMLSBert) — Xenova/UMLSBert_ENG, 111 MB, 109M params
• MedEmbed Small — Romelianism/MedEmbed-small-v0.1, 35 MB, 33M params

Entity Recognition

• BERT Base NER — Xenova/bert-base-NER, 110 MB, 108M params
• mBERT NER (10 languages) — Xenova/bert-base-multilingual-cased-ner-hrl, 182 MB, 177M params
• DistilBERT NER (10 languages) — Xenova/distilbert-base-multilingual-cased-ner-hrl, 139 MB, 135M params
• Piiranha v1 (PII) — onnx-community/piiranha-v1-detect-personal-information-ONNX, 334 MB, 280M params
• BERT Small PII — onnx-community/bert-small-pii-detection-ONNX, 30 MB
• DistilBERT NER — onnx-community/distilbert-NER-ONNX, 67 MB, 65M params
• OpenMed PII Small 44M — YashaBor/OpenMed-PII-SuperClinical-Small-44M-v1-ONNX, 181 MB, 44M params
• Biomedical NER (84 types) — onnx-community/biomedical-ner-all-ONNX, 68 MB, 65M params
• Medical NER (41 types) — onnx-community/Medical-NER-ONNX, 231 MB, 184M params
• Stanford De-identifier — onnx-community/stanford-deidentifier-base-ONNX, 111 MB, 108M params
• OpenMed Disease NER — onnx-community/OpenMed-NER-DiseaseDetect-BioMed-335M-ONNX, 337 MB, 335M params
• OpenMed PII Base 184M — onnx-community/OpenMed-PII-SuperClinical-Base-184M-v1-ONNX, 252 MB, 184M params
• OpenMed Drug NER — onnx-community/OpenMed-NER-PharmaDetect-SuperMedical-125M-ONNX, 130 MB, 125M params
• Italian Medical NER — onnx-community/italian_medical_ner-ONNX, 109 MB, 108M params

Depth Estimation

• Depth Anything V2 Small — onnx-community/depth-anything-v2-small, 99 MB, 24.8M params
• Depth Anything V1 Small — Xenova/depth-anything-small-hf, 99 MB, 24.8M params
• MiDaS DPT Hybrid — Xenova/dpt-hybrid-midas, 533 MB, 123M params
• Depth Anything V2 Base — onnx-community/depth-anything-v2-base, 389 MB, 97.5M params
• GLPN NYU — Xenova/glpn-nyu, 246 MB, 62M params
• Metric3D ViT Small — onnx-community/metric3d-vit-small, 151 MB

Find Anything

• OWL-ViT Base 32 — Xenova/owlvit-base-patch32, 159 MB, 154M params
• Grounding DINO Tiny — onnx-community/grounding-dino-tiny-ONNX, 205 MB
• OWLv2 Base 16 Ensemble — Xenova/owlv2-base-patch16-ensemble, 159 MB
• OWL-ViT Base 16 — Xenova/owlvit-base-patch16, 159 MB, 154M params
• OWL-ViT Large 14 — Xenova/owlvit-large-patch14, 441 MB, 430M params

Text to Speech

• Kokoro 82M v1.0 — onnx-community/Kokoro-82M-v1.0-ONNX, 93 MB, 82M params
• Kokoro 82M v0.19 — onnx-community/Kokoro-82M-ONNX, 92 MB, 82M params
• SpeechT5 TTS — Xenova/speecht5_tts, 178 MB, 144M params
• MMS TTS English — Xenova/mms-tts-eng, 114 MB, 36M params
• Kokoro 82M v1.1 (Chinese) — onnx-community/Kokoro-82M-v1.1-zh-ONNX, 127 MB, 82M params
• MMS TTS Hindi — Xenova/mms-tts-hin, 114 MB, 36M params
• MMS TTS Spanish — Xenova/mms-tts-spa, 114 MB, 36M params

Question Answering

• DistilBERT SQuAD (cased) — Xenova/distilbert-base-cased-distilled-squad, 67 MB, 65M params
• DistilBERT SQuAD (uncased) — Xenova/distilbert-base-uncased-distilled-squad, 68 MB, 66M params
• ModernBERT SQuAD — onnx-community/ModernBERT-finetuned-squad-ONNX, 338 MB, 150M params
• MobileBERT SQuAD2 — onnx-community/mobilebert-uncased-squad-v2-ONNX, 28 MB, 25M params
• RoBERTa SQuAD2 — onnx-community/roberta-base-squad2-ONNX, 130 MB, 125M params
• TinyRoBERTa SQuAD2 — onnx-community/tinyroberta-squad2-ONNX, 87 MB

Speech to Text

• Whisper Tiny (English) — Xenova/whisper-tiny.en, 67 MB, 39M params
• Whisper Tiny (multilingual) — Xenova/whisper-tiny, 68 MB, 39M params
• Whisper Base (Xenova) — Xenova/whisper-base, 141 MB, 74M params
• Whisper Small — onnx-community/whisper-small, 514 MB, 244M params
• Whisper Base — onnx-community/whisper-base, 141 MB, 74M params
• Whisper Large v3 Turbo — onnx-community/whisper-large-v3-turbo, 983 MB, 809M params
• Moonshine Tiny — onnx-community/moonshine-tiny-ONNX, 55 MB, 27M params

Object Detection

• DETR ResNet-50 — Xenova/detr-resnet-50, 167 MB, 41M params
• YOLOS Tiny — Xenova/yolos-tiny, 26 MB, 6.5M params
• RF-DETR Nano — onnx-community/rfdetr_nano-ONNX, 108 MB
• YOLOS Small — Xenova/yolos-small, 123 MB, 30.7M params
• DETR ResNet-101 — Xenova/detr-resnet-101, 243 MB, 60M params
• Table Transformer — Xenova/table-transformer-detection, 116 MB, 28.8M params

Audio Classification

• AST AudioSet — Xenova/ast-finetuned-audioset-10-10-0.4593, 91 MB, 86M params
• Common Voice Gender Detection — prithivMLmods/Common-Voice-Gender-Detection-ONNX, 95 MB, 94M params
• wav2vec2 Speech Emotion — onnx-community/wav2vec2-base-Speech_Emotion_Recognition-ONNX, 95 MB, 94M params
• MMS Language ID (126) — Xenova/mms-lid-126, 974 MB, 965M params
• AST Speech Commands — Xenova/ast-finetuned-speech-commands-v2, 87 MB, 86M params
• wav2vec2 Keyword Spotting — Xenova/wav2vec2-base-superb-ks, 96 MB, 94M params

Fill-Mask

• BERT Base Uncased — Xenova/bert-base-uncased, 112 MB, 110M params
• DistilBERT Base Uncased — Xenova/distilbert-base-uncased, 69 MB, 67M params
• ALBERT Base v2 — Xenova/albert-base-v2, 43 MB, 12M params
• mBERT Cased — Xenova/bert-base-multilingual-cased, 184 MB, 178M params
• Bio_ClinicalBERT — Xenova/Bio_ClinicalBERT, 111 MB, 108M params
• RoBERTa Base — Xenova/roberta-base, 130 MB, 125M params
• DistilRoBERTa Base — Xenova/distilroberta-base, 86 MB, 82M params

Summarization

• DistilBART CNN 6-6 — Xenova/distilbart-cnn-6-6, 287 MB, 230M params
• BART Large CNN — Xenova/bart-large-cnn, 466 MB, 406M params
• DistilBART CNN 12-6 — Xenova/distilbart-cnn-12-6, 364 MB, 306M params
• DistilBART XSum 6-6 — Xenova/distilbart-xsum-6-6, 287 MB, 230M params
• DistilBART XSum 12-6 — Xenova/distilbart-xsum-12-6, 364 MB, 306M params
• DistilBART CNN 12-3 — Xenova/distilbart-cnn-12-3, 312 MB, 255M params

Zero-Shot Text Labels

• BART Large MNLI — Xenova/bart-large-mnli, 415 MB, 407M params
• MobileBERT MNLI — Xenova/mobilebert-uncased-mnli, 28 MB, 25M params
• DeBERTa v3 XSmall NLI — Xenova/nli-deberta-v3-xsmall, 98 MB, 22M params
• DistilBERT MNLI — Xenova/distilbert-base-uncased-mnli, 69 MB, 67M params
• mDeBERTa v3 XNLI — Xenova/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7, 359 MB, 279M params
• DeBERTa v3 Small NLI — Xenova/nli-deberta-v3-small, 184 MB, 44M params

Sentiment

• Twitter RoBERTa Sentiment — Xenova/twitter-roberta-base-sentiment-latest, 129 MB, 125M params
• DistilBERT SST-2 — Xenova/distilbert-base-uncased-finetuned-sst-2-english, 69 MB, 67M params
• Toxic BERT (Detoxify) — Xenova/toxic-bert, 112 MB, 109M params
• mBERT Star Rating — Xenova/bert-base-multilingual-uncased-sentiment, 172 MB, 167M params
• FinBERT — Xenova/finbert, 112 MB, 109M params
• Multilingual Sentiment Student — Xenova/distilbert-base-multilingual-cased-sentiments-student, 140 MB, 135M params

Teach a Classifier

• DINOv2 Small — Xenova/dinov2-small, 25 MB, 22.1M params

Translation

• NLLB-200 Distilled 600M — Xenova/nllb-200-distilled-600M, 917 MB, 600M params
• Marian EN → ZH — Xenova/opus-mt-en-zh, 123 MB, 77M params
• Marian EN → ES — Xenova/opus-mt-en-es, 123 MB, 77M params
• Marian EN → FR — Xenova/opus-mt-en-fr, 116 MB, 75M params
• Marian EN → DE — Xenova/opus-mt-en-de, 115 MB, 74M params
• Marian Multi → EN — Xenova/opus-mt-mul-en, 121 MB, 77M params
• Marian EN → HI — Xenova/opus-mt-en-hi, 121 MB, 76M params

OCR

• TrOCR Small (printed) — Xenova/trocr-small-printed, 70 MB, 62M params
• TrOCR Small (handwritten) — Xenova/trocr-small-handwritten, 70 MB, 62M params
• TrOCR Base (handwritten) — Xenova/trocr-base-handwritten, 341 MB, 334M params
• TrOCR Base (printed) — Xenova/trocr-base-printed, 341 MB, 334M params
• Nougat Small — Xenova/nougat-small, 258 MB, 250M params
• Texify — Xenova/texify, 321 MB