AlibabaQwen Guard
Qwen3Guard-Stream-0.6B
Qwen3Guard-Stream adds token-level classification for incremental text. It supports 119 languages and dialects.
Practical use
Practical guidance. Measure detection delay in tokens as well as classification accuracy. Decide how much text can be buffered before display so a late flag does not arrive after the content has already been shown.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | Qwen/Qwen3Guard-Stream-0.6B |
| Training family | The Guard series uses 1.19 million labelled prompts and responses.Qwen3Guard-Stream-0.6B model card |
| Interface | Streaming token-level safety signals.Qwen3Guard-Stream-0.6B model card |
| Licence | Apache 2.0.Qwen3Guard-Stream-0.6B model card |
Related models
- Qwen3Guard-Gen-0.6B
Alibaba · Qwen Guard
A 600-million-parameter Qwen3 moderation model for complete text in 119 languages and dialects.
Launch date not confirmedDownloadable - Qwen3Guard-Gen-4B
Alibaba · Qwen Guard
Four-billion-parameter generative moderation model trained from Qwen3 for 119 languages and dialects.
Launch date not confirmedDownloadable - Qwen3Guard-Gen-8B
Alibaba · Qwen Guard
Eight-billion-parameter Qwen3-based model that treats text moderation as an instruction task.
Launch date not confirmedDownloadable