Text Classification
Transformers
Safetensors
modernbert
feature-extraction
profanity
offensive-language
hate-speech
content-moderation
indic
multilingual
code-mixed
multi-label
text-embeddings-inference
Instructions to use CobuildX-AI/cbx-indic-profanity-detector-36M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use CobuildX-AI/cbx-indic-profanity-detector-36M with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="CobuildX-AI/cbx-indic-profanity-detector-36M")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("CobuildX-AI/cbx-indic-profanity-detector-36M") model = AutoModel.from_pretrained("CobuildX-AI/cbx-indic-profanity-detector-36M", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download NOTICE from CobuildX-AI/cbx-indic-profanity-detector-36M: direct link, hf CLI and curl.
- Browser
- Download file 3.76 kB
-
https://huggingface.co/CobuildX-AI/cbx-indic-profanity-detector-36M/resolve/main/NOTICE
- Command line
-
hf download hf://CobuildX-AI/cbx-indic-profanity-detector-36M/NOTICE
-
curl -L -o NOTICE https://huggingface.co/CobuildX-AI/cbx-indic-profanity-detector-36M/resolve/main/NOTICE
3.76 kB
| cbx-indic-profanity-detector-36M | |
| Copyright 2026 Cobuild Tech Private Limited. | |
| Licensed under the Apache License, Version 2.0 (see LICENSE): the model weights, configuration, tokenizer, scripts | |
| and documentation in this repository. Provided "as is", without warranties or conditions of any kind (LICENSE, | |
| sections 7 and 8); see "Intended use and disclaimer" in README.md. | |
| This model contains or was built from the following third-party materials: | |
| 1. mmBERT (jhu-clsp/mmBERT-small, revision abc32620dd4f6ab06f5fbe905dc25f310618e09f), the base model (its vocabulary was pruned and, for the smaller | |
| sizes, layers removed). Copyright (c) 2025 the mmBERT authors (Marone, Weller, Fleshman, Yang, Lawrie, | |
| Van Durme). MIT License: | |
| Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated | |
| documentation files (the "Software"), to deal in the Software without restriction, including without limitation the | |
| rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit | |
| persons to whom the Software is furnished to do so, subject to the following conditions: | |
| The above copyright notice and this permission notice shall be included in all copies or substantial portions of the | |
| Software. | |
| THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE | |
| WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR | |
| COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR | |
| OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. | |
| 2. Training text (not distributed in this repository). The model was trained on text from these datasets, whose | |
| authors we thank and credit: | |
| - MACD (Multilingual Abusive Comment Dataset), ShareChat / Moj comments, ShareChat (hi, ta, te, ml, kn): https://github.com/ShareChatAI/MACD. Licence: no licence file in the repository; used with attribution. | |
| - DravidianCodeMix offensive language identification (OffensEval Dravidian), Chakravarthi et al. (ta, ml, kn): https://huggingface.co/datasets/community-datasets/offenseval_dravidian. Licence: CC BY 4.0. | |
| - BD-SHS, Romim et al., LREC 2022 (bn): https://www.kaggle.com/datasets/naurosromim/bdshs. Licence: CC BY 4.0. | |
| - Bengali Hate Speech, Karim et al. (bn): https://huggingface.co/datasets/rezacsedu/bn_hate_speech. Licence: MIT. | |
| - Telugu-Hatespeech, Marreddy et al., ACM TALLIP 2022 (te): https://huggingface.co/datasets/mounikaiiith/Telugu-Hatespeech. Licence: CC BY 4.0. | |
| - MOLD (Marathi Offensive Language Dataset) 1.0 and 2.0, Gaikwad, Zampieri et al. (mr): https://github.com/tharindudr/MOLD. Licence: CC BY 4.0. | |
| - RUHSOLD (Roman Urdu Hate-Speech and Offensive Language Dataset), Rizwan et al., EMNLP 2020 (ur (romanized)): https://github.com/haroonshakeel/roman_urdu_hate_speech. Licence: MIT. | |
| - Hate Speech and Offensive Language, Davidson et al., ICWSM 2017 (en): https://github.com/t-davidson/hate-speech-and-offensive-language. Licence: MIT. | |
| - HateXplain, Mathew et al., AAAI 2021 (en): https://github.com/hate-alert/HateXplain. Licence: CC BY 4.0. | |
| - FineWeb-2 (web text), Hugging Face (all 12 Indic languages): https://huggingface.co/datasets/HuggingFaceFW/fineweb-2. Licence: ODC-By 1.0. | |
| No dataset text is distributed with the model. Use of the model does not imply endorsement by any of these | |
| authors or publishers. | |
| 3. Labels. Part of the training labels were produced by large language models: Gemma 4 31B (Apache-2.0), | |
| Claude Haiku 5.5 and Claude Sonnet 5.5 (Anthropic), and gpt-5.4-mini (OpenAI). | |