Traefik Hub's AI Gateway now supports a multi-vendor safety pipeline, allowing organizations to choose from multiple guardrail providers and combine them.
Request early access to Traefik Hub v3.20 and start building AI security pipelines with:
- Regex Guard: A framework for organizations to write their own guards using regex-based pattern matching.
- Content Guard (Microsoft Presidio): PII detection and masking with statistical NLP-based entity recognition, supporting both built-in and custom entity patterns.
- LLM Guard with NVIDIA NIMs: GPU-accelerated jailbreak detection, content safety across 22+ categories, and topic control.
- LLM Guard with IBM Granite Guardian: IBM's open-source safety models provide harm detection, jailbreak detection, topic control, hallucination detection, and RAG quality assessment, capabilities that other guard providers don't yet offer.
- Parallel Guard Execution: LLM-based guards, the heavyweight tiers that can take multiple seconds to execute, now run in parallel rather than in series.
Request early access today and layer your AI security without increasing latency.
