Content Moderation Qwen3 0.6B CoT · 73% accuracy

Malaysia Content Moderator

Analyze Malaysian text for harmful content using a fine-tuned language model with chain-of-thought reasoning. Detects communal hostility, religious provocation, misinformation, and more.

Input Text
0 chars
Try an example: Safe — everyday chat Safe — quoting to condemn Unsafe — communal hostility Unsafe — political sarcasm Safe — pro-social defense Unsafe — coded euphemism
Analyzing content with chain-of-thought reasoning...