OrcaRouter Has Released OrcaCyber Zero 1.5
The new cybersecurity model features a 1 million-token context window and expanded support for automated vulnerability testing.
Updated on Oct. 11, 2026 in Cybersecurity
Live Poll
Do you trust that advanced AI models for vulnerability research will make the internet more secure?
OrcaRouter has launched the OrcaCyber Zero 1.5 model, an upgraded iteration designed for exploit development and penetration testing. The model supports native function calling and includes a 1 million-token context window for complex security tasks.
Why it matters
This model provides advanced capabilities for vulnerability reproduction and security research. By automating complex workflows, it assists security professionals in identifying and addressing potential exploits more effectively.
OrcaCyber Zero 1.5 features a 1 million-token context window and supports structured outputs via an OpenAI-compatible API. The model operates with input token pricing of $3.00 and output pricing of $7.50 per 1 million tokens.
The players
OrcaRouter
This technology company hosts the platform and API infrastructure through which the OrcaCyber model series is accessed.
The details
Access to the model is restricted to the Security Research tier, requiring users to complete an engagement process and provide a passkey. The platform reports a p50 time-to-first-token latency of 500 milliseconds for the new system.
Timeline
OrcaCyber Zero 1.0 was released on September 17, 2026.
Model performance benchmarks were evaluated on October 10, 2026.
The official release article for version 1.5 was published on October 11, 2026.
The Tech Race
The introduction of version 1.5 marks an evolution in AI-driven cybersecurity tools, moving beyond simple classification toward automated vulnerability reproduction. This release positions OrcaRouter to compete with emerging specialized agents currently replacing legacy manual penetration testing workflows.
Professional security researchers can now utilize native function calling and a massive context window to streamline complex exploit development workflows. Users must navigate an approval process involving a passkey to access these capabilities via the hosted API.
The takeaway
This release highlights the rapid integration of large-scale context windows into specialized security research tools. Practitioners should note that access remains gated to ensure specialized usage protocols are followed.
Further reading
For broader trends in AI-driven safety tools, visit Cybersecurity.
Source note: This article includes information reported by MarkTechPost.
Live Poll
Do you trust that advanced AI models for vulnerability research will make the internet more secure?
