# Qwen3Guard: first safety guardrail model in Qwen family

> Qwen released Qwen3Guard, a safety classification model built on Qwen3 and fine-tuned for prompt and response moderation. It outputs risk levels and category classifications, and claims state-of-the-art results on major safety benchmarks per the vendor's tech report.

| | |
|---|---|
| **Tool** | Qwen |
| **Version** | — |
| **Kind** | announcement |
| **Published** | 2025-09-22 |
| **Observed** | 2026-08-11 |
| **Significance** | 3/5 |
| **Breaking** | no |
| **Categories** | new-model, capability, feature, security |

## What changed


- First safety guardrail model in the Qwen family
- Built on Qwen3, fine-tuned for safety classification
- Detects risks in both prompts and responses
- Outputs risk levels and category classifications
- Supports English, Chinese, and multilingual
- Claims state-of-the-art on major safety benchmarks
- Published as tech report, GitHub repo, Hugging Face model, and ModelScope


## Sources

- [blog_rss](https://qwenlm.github.io/blog/qwen3guard/) — retrieved 2026-08-11


## Community

_No curated reactions recorded. Facts and community takes are kept in separate layers
and never blended._

---
Canonical: https://changelogs.info/qwen/qwen3guard-first-safety-guardrail-model-in-qwen-family
Entity: https://changelogs.info/qwen
Event ID: `evt_2025-09-22_qwen_qwen3guard-real-time-safety-for-your-token-stream`
Licence: event synthesis © changelogs.info, CC BY 4.0. Linked sources belong to their vendors.