Antares is a family of security-focused small language models (SLMs) developed by Cisco for the specific task of localizing known vulnerabilities within source code. The models are open-weight, highly efficient, and designed to reduce the time and cost associated with traditional security auditing processes. They are intended for integration into developer and security workflows to automatically identify vulnerable code sections.
Antares is an AI & ML project. Manually locating known vulnerabilities in large codebases is slow, expensive, and error-prone. Antares is an open-source project aimed at security engineers and developers. Antares is open source under the Open Source license. It ships for the web, the command line, and API.
Behind Antares is Cisco, based in the United States, and it first shipped in 2026. Among its 4 catalogued features are Vulnerability Localization, Small Language Models, and Code Analysis.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match