Skip to content

AI Security Research Lab

Kanti Labs logo KANTILABS

AI Security Research and Offensive Security Models

Kanti Labs is an AI security research lab building open-weight models, datasets, and research for offensive web security.

4
Models Released
500+
Model Downloads
1
Open Source Tool
2
Research Posts

Kanti Labs Models

Open-weight offensive security models fine-tuned for web exploitation research, local inference, and Strix-based workflows.

SFT

kanti-xss-sft-4b

Supervised fine-tuned on curated XSS synthetic Strix traces from real vulnerabilities. Designed to be used as a Strix sub-agent.

BaseQwen3-4B
MethodSupervised Fine-Tuning
TaskXSS Payload Generation
RL

kanti-xss-rl-4b

Reinforcement learning variant trained with reward signals from simulated XSS validation. Designed to be used as a Strix sub-agent.

BaseQwen3-4B
MethodReinforcement Learning
TaskXSS Payload Generation

Research Roadmap

Where Kanti Labs has been and what the next model and research releases are aiming toward.

Complete Phase 1

XSS Foundation Models

Released SFT and RL 4B-parameter models for XSS payload generation. Established training pipeline, dataset curation, and evaluation methodology.

In Progress Phase 2

Full Post-Training Run

Completing comprehensive post-training with expanded datasets and improved reward modeling. The next Kanti Labs model release is the near-term milestone.

Planned Phase 3

Scale and Expand

Scale to larger dense and MoE architectures, then expand beyond XSS into SQLi, SSRF, and adjacent vulnerability classes.

Future Phase 4

Unified Multi-Vulnerability Model

A single model capable of reasoning about multiple vulnerability families with tighter integration into defense evaluation loops.

Contact Kanti Labs

Kanti Labs is available for research collaboration, sponsorship discussions, and conversations about offensive security models.