{"slug": "cisco-bets-its-small-open-cybersecurity-models-can-outperform-gpt-5-5-at-for-a", "title": "Cisco bets its small open cybersecurity models can outperform GPT-5.5 at vulnerability detection for a fraction of the cost", "summary": "Cisco released two small open-source AI models, Antares-350M and Antares-1B, that detect software vulnerabilities at a fraction of the cost of larger models. In Cisco's tests, Antares scanned 500 code repositories in about 15 minutes for under a dollar, while GPT-5.5 took five hours and cost over $100 for the same job. Developer Aman Priyanshu claimed the smallest model catches about 150 times more vulnerabilities per dollar than large AI agents like Cognition's Devin Security Swarm.", "body_md": "# Cisco bets its small open cybersecurity models can outperform GPT-5.5 at vulnerability detection for a fraction of the cost\n\n**Cisco has released two small, open AI models for cybersecurity.** [Antares-350M and Antares-1B](https://huggingface.co/collections/fdtn-ai/antares) spot vulnerabilities in software code. Cisco's pitch is bang for the buck. [Developer Aman Priyanshu claims on X](https://x.com/AmanPriyanshu6/status/2079703080640364848) that the smallest model catches about 150 times more vulnerabilities per dollar than large AI agents like Cognition's Devin Security Swarm. In Cisco's own tests, [according to Axios](https://www.axios.com/2026/07/21/cisco-open-source-ai-models-cybersecurity), Antares scanned 500 code repos in about 15 minutes for under a dollar. GPT-5.5 took five hours and cost over $100 for the same job.\n\nBoth models run locally, so sensitive code never leaves the company. [The technical report](http://cisco-foundation-ai.github.io/antares/technical-report.pdf) says the models were trained on roughly 72 percent security-concept data and 15 percent code search histories.\n\nCisco is keeping a larger three-billion-parameter version for its own products. That version reportedly performs close to GPT-5.5 and beats open models up to 200 times its size. The company is also exploring an industry consortium for open AI security tools.\n\n```\nAI News Without the Hype – Curated by Humans\n\n\t\t\t\t\tSubscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive \"AI Radar\" frontier report six times a year, full archive access, and access to our comment section.\t\t\t\t\n\n\t\t\t\t\tSubscribe now\n```\n\n[Paper](https://cisco-foundation-ai.github.io/antares/technical-report.pdf)|\n\n[Hugging Face](https://huggingface.co/collections/fdtn-ai/antares)|\n\n[via X](https://twitter-thread.com/t/2079703080640364848)|\n\n[via Axios](https://www.axios.com/2026/07/21/cisco-open-source-ai-models-cybersecurity)", "url": "https://wpnews.pro/news/cisco-bets-its-small-open-cybersecurity-models-can-outperform-gpt-5-5-at-for-a", "canonical_source": "https://the-decoder.com/cisco-bets-its-small-open-cybersecurity-models-can-outperform-gpt-5-5-at-vulnerability-detection-for-a-fraction-of-the-cost/", "published_at": "2026-07-22 16:28:08+00:00", "updated_at": "2026-07-22 16:56:29.333115+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-products", "ai-tools", "ai-safety", "ai-research"], "entities": ["Cisco", "Antares-350M", "Antares-1B", "GPT-5.5", "Aman Priyanshu", "Cognition", "Devin Security Swarm", "Axios"], "alternates": {"html": "https://wpnews.pro/news/cisco-bets-its-small-open-cybersecurity-models-can-outperform-gpt-5-5-at-for-a", "markdown": "https://wpnews.pro/news/cisco-bets-its-small-open-cybersecurity-models-can-outperform-gpt-5-5-at-for-a.md", "text": "https://wpnews.pro/news/cisco-bets-its-small-open-cybersecurity-models-can-outperform-gpt-5-5-at-for-a.txt", "jsonld": "https://wpnews.pro/news/cisco-bets-its-small-open-cybersecurity-models-can-outperform-gpt-5-5-at-for-a.jsonld"}}