Explore the latest research papers, architectural releases, sovereign intelligence milestones, and announcements from Cortiqa.
Announcing the research preview of Falin-300M, a 297M parameter decoder-only SLM engineered from scratch on our proprietary Menothus architecture. Designed with 8:1 Grouped-Query Attention and Parallel SwiGLU FFN for ultra-low latency inference on sovereign and edge hardware.
Read announcementOur latest research on why brute-force context expansion fails, and how Hierarchical Semantic State Tracking achieves infinite conversational continuity without token bloat.
Public release of Kricko, our 100% offline LLM chat application running models directly on your device. Zero data collection, complete privacy, and zero internet dependency.
Integration of Falin-01, our sovereign conversational model optimized for real-time dialogue and local inference with high efficiency.
Get platform announcements, model releases, and research papers direct in your inbox.
For press inquiries, interviews, or media kit requests, reach out to our communications team.