cointegrated / text-classification Open weights updated 3 months ago

rubert-tiny-toxicity

Name: rubert-tiny-toxicity
Author: cointegrated

This is the cointegrated/rubert-tiny model fine-tuned for classification of toxicity and inappropriateness for short informal Russian texts, such as comments in social networks.

Params

10 M

Context

512

Downloads 30d

258 K

Likes

Commercial use: allowed mit Not gated SAFETENSORS 1 languages View on Hugging Face ↗

Download history

daily snapshots · 79 days

▲ 0 in the last 30 days (0.0%)

452 K258 K

Aug 24Sep 19Oct 15Nov 10

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	f32	0.0 GB	0.6 GB	✅ Runs comfortably

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Specifications

Architecture: BertForSequenceClassification
Parameters: 10 M
Tensor type: F32
Context length: 512
Vocabulary: 29,564
Layers / heads: 3 / 12
Licence: mit
First seen on the Hub: 2022-03-02
Training datasets: undisclosed
Added to our catalog: 2026-08-24

Compare with any text-classification model

Popular comparisons

twitter-roberta-base-sentiment-latest 2.4 M fasttext-language-identification 830 K ms-marco-MiniLM-L-6-v2 710 K glotlid 641 K twitter-xlm-roberta-base-sentiment 613 K finbert-tone 557 K