Paper: SciFive: a text-to-text transformer model for biomedical literature Authors: Long N. Phan, James T. Anibal, Hieu Tran, Shaurya Chanana, Erol Bahadroglu, Alec Peltekian, Grégoire Altan-Bonnet For more details, do check out our Github repo.
Open-weight model · Text classification
muril-cyberbullying-detection
by Suyash Sahu suyashsahu00/muril-cyberbullying-detection
muril-cyberbullying-detection is an open-weight model for text classification from Suyash Sahu, released under MIT License. It has 238M parameters and a 512-token context. At 16-bit it needs about 0.6 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 31 downloads a month.
This model is a fine-tuned version of Google's MuRIL (Multilingual Representations for Indian Languages) BERT architecture, specifically adapted for multi-class Cyberbullying & Hate Speech Detection.
Parameters238M
Context512
Weights950.3 MB
Licensemit
AccessOpen weights
Monthly Downloads31
Runs On
What it takes to serve muril-cyberbullying-detection (238M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 0.5 GB | 0.6 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.2 GB | 0.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 7, 2026.
muril-cyberbullying-detection on every accelerator the SAVRN Index prices, at every precision
Model Card
By Suyash Sahu, published under mit, revision e8a34004c520.
MuRIL Multilingual Cyberbullying Detection (v2)
This model is a fine-tuned version of Google's MuRIL (Multilingual Representations for Indian Languages) BERT architecture, specifically adapted for multi-class Cyberbullying & Hate Speech Detection.
It categorizes social media text and online commentary across 6 categories:
1. age: Cyberbullying targeting an individual's age.
2. ethnicity: Bullying or hate speech targeting ethnicity, race, or caste.
3. gender: Misogyny, sexism, or gender-based harassment.
4. religion: Hate speech or insults targeting religious beliefs.
5. other_cyberbullying: General toxic harassment, insults, or threats.
6. not_cyberbullying: Benign, safe, neutral, or positive text.
Model Performance
Evaluated on the held-out multi-class test benchmark:
| Metric | Score |
|---|---|
| Overall Accuracy | 81.97% |
| Macro Precision | 83.21% |
| Macro Recall | 83.41% |
| Macro F1-Score | 83.29% |
Per-Class F1 Breakdown:
- Age: 97.76% F1
- Ethnicity: 95.86% F1
- Religion: 95.03% F1
- Gender: 86.32% F1
- Not Cyberbullying: 64.63% F1
- Other Cyberbullying: 60.11% F1
Quickstart & Inference
Configuration
- Architecture
- BertForSequenceClassification
- Context length (tokens)
- 512
- Layers
- 12
- Hidden size
- 768
- Feed-forward size
- 3,072
- Attention heads
- 12
- Vocabulary size
- 197,285
- Model type
- bert
Identity and Version
- Repository
- suyashsahu00/muril-cyberbullying-detection
- Publisher
- Suyash Sahu
- Task
- Text classification
- Modality
- Text
- Library
- Not stated by the source
- Parameters
- 238M parameters
- Languages
- en, hi
- Revision
- e8a34004c520b7363f70e19e6e4a0aae1b51072f
- First published
- 2026-09-03
- Last updated
- 2026-09-23
Files and Weights
7 files, 956.7 MB in total. The weights are 1 file totalling 950.3 MB in safetensors.
Weights1 file · 950.3 MB
Configuration2 files · 1.4 KB
Tokenizer2 files · 6.4 MB
Documentation1 file · 3.1 KB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 950.3 MB | 0b671e02716a |
| config.json | Configuration | 1.1 KB | — |
| label_map.json | Configuration | 331 B | — |
| README.md | Documentation | 3.1 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| tokenizer.json | Tokenizer | 6.4 MB | — |
| tokenizer_config.json | Tokenizer | 390 B | — |
License and Download
- License
- mit
- Access
- Open weights, no gate
- Download size
- 950.3 MB
Download from Suyash Sahu
Released by Suyash Sahu through its official repository on Hugging Face. Read the license.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 950.3 MB |
| 16-bit | 0.5 GB |
| 8-bit | 0.2 GB |
| 4-bit | 0.1 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About muril-cyberbullying-detection
How much GPU memory does muril-cyberbullying-detection need?
About 0.6 GB at 16-bit and 0.1 GB at 4-bit: the weights (238M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run muril-cyberbullying-detection on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use muril-cyberbullying-detection commercially?
Yes. muril-cyberbullying-detection is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.
What is muril-cyberbullying-detection's context length?
512 tokens, from the maximum position embeddings in its published configuration.
Similar Models
A sequence classification model fine-tuned on top of ai4bharat/IndicBERTv2-MLM-only to detect fraudulent, phishing, and scam messages across 14 Indian languages and language varieties. The model is designed with a focus on lightweight multilingual scam detection for Indian communications and common fraud patterns. This model is intended to detect common fraud and scam patterns prevalent across Indian communications, including: - Electricity and utility disconnection threats. - Impersonation of major institutions such as banks, India Post, and courier services. - Fake lottery, subsidy, and government-scheme claims. - Suspicious payment requests and fee demands. - Phishing and malicious…
A sequence classification model fine-tuned on top of ai4bharat/IndicBERTv2-MLM-only to detect fraudulent, phishing, and scam messages across 14 Indian languages and language varieties[cite: 4]. This v2 model represents a significant upgrade over the baseline v1 iteration, leveraging advanced entity masking and continuous feedback loop training on a T4 GPU to heavily reduce False Negatives (scam $ightarrow$ ham misclassifications). The original v1 baseline achieved a highly respectable 98.29% accuracy but exhibited vulnerabilities to specific scam evasion tactics (e.g., protocol obfuscation like hxxp://, naked domains, and specific tele-fraud requests)[cite: 4]. To resolve this, v2 was…
We have updated the new reranker, supporting larger lengths, more languages, and achieving better performance. More details please refer to our Github: FlagEmbedding. FlagEmbedding focuses on retrieval-augmented LLMs, consisting of the following projects currently: - 3/18/2024: Release new rerankers, built upon powerful M3 and LLM (GEMMA and MiniCPM, not so large actually) backbones, supporitng multi-lingual processing and larger inputs, massive improvements of ranking performances on BEIR, C-MTEB/Retrieval, MIRACL, LlamaIndex Evaluation. - 3/18/2024: Release Visualized-BGE, equipping BGE with visual capabilities. Visualized-BGE can be utilized to generate embeddings for hybrid image-text…
This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 3e-05 - trainbatchsize: 16 - evalbatchsize: 16 - lrschedulertype: linear - trainingsteps: 1800 - mixedprecisiontraining: Native AMP - Transformers 4.37.2 - Pytorch 2.1.0+cu121 - Datasets 2.17.1 - Tokenizers 0.15.2
This repository contains a fine-tuned XLM-RoBERTa Base model for 3-class sentiment classification in Sindhi. The model predicts one of the following labels: - positive - neutral - negative This model is intended for research, experimentation, and lightweight production prototyping on Sindhi-language sentiment analysis tasks. It may be useful for short-text classification such as reviews, feedback, and social-media snippets. It should not be used as the only basis for high-stakes decisions. Performance can vary by domain, dialect, and writing style, and human review is recommended when errors could affect people or organizations. The saved model artifacts in this repository report the…