OrcaCyber Zero 1.5

OrcaCyber Zero 1.5 is a frontier cybersecurity model and the successor to OrcaCyber Zero 1.0, post-trained for security research and authorized security engineering: vulnerability research and reproduction, exploit development, penetration testing, security auditing, and advanced cyber reasoning. It is built for complex, multi-step security workflows, with long-context reasoning, native function calling, and structured outputs. Vendor-reported evaluations: 95.8% on the evaluable CVE-Bench subset, 100% on Cybench (unrestricted agent execution), 76.5% on SWE-bench Pro V2, and 93.9% on HumanEval+. Like Zero 1.0 it is served through an OpenAI-compatible API and is access-gated to the Security Research tier for trusted security researchers, red teams, and authorized security testing.