Compare vendors / garak
garak
Open-source LLM vulnerability scanner
Agents and toolsCompliance mappingOpen source
garak probes LLMs for prompt injection, jailbreaks, data leakage, hallucination, misinformation and toxicity using probes and detectors. Probe tags map to OWASP LLM Top 10, and v0.17.0 (Sep 2026) added EU AI Act risk category tags. It has about 10,000 GitHub stars, and v0.17.1 shipped on Oct 16, 2026.
Best for: A free first scan of an LLM endpoint for known weaknesses.
Open source (Apache-2.0), maintained by NVIDIA
- ✓ Maps to OWASP
- – Maps to MITRE ATLAS
- ✓ Maps to the EU AI Act
- – SOC 2 or ISO 27001
1 Legal namegarak
2 HeadquartersNot disclosed
3 Founded2023
4 TeamNot disclosed
5 What they testLLM apps, Prompt injection and jailbreaks, Data leakage, Bias and harms, Agents and tool use
6 FrameworksOWASP LLM Top 10, EU AI Act, AVID effect taxonomy, CWE, Language Model Risk Cards
7 CertificationsNot disclosed
8 Published pricesFree (Apache-2.0)
9 Customers and partnersNot disclosed
Where we recommend them
- Best open-source tool, our pick.
Sources
- GitHub: NVIDIA/garak
- GitHub: releases
- arXiv: garak paper
- NVIDIA NeMo Guardrails docs: scanning with garak
Work at garak? Send corrections with a source to editors@bestairedteamingcompanies.com. We fix verified errors within five business days.
Compare with
- Mindgard AI red teaming firm: human-led tests plus an automated platform
- Gray Swan AI AI security lab: agent red teaming, lab evals and the Arena
- HiddenLayer AI security platform with automated red teaming and model scanning
- Trail of Bits Security research consultancy with an AI/ML assurance practice
- Pillar Security Agent security platform with agentic red teaming