July 20, 2026

RAGU is an open-source GraphRAG system that separates entity extraction from consolidation via a six-stage pipeline (typed extraction, DBSCAN deduplication, community detection), paired with Meno-Lite-0.1, a 7B fine-tuned extraction model that outperforms Qwen2.5-32B on knowledge-graph construction by 12.5%.

xHC extends Hyper-Connections beyond N=4 parallel residual streams by combining causal-convolution write-back augmentation with sparse residual updates (activating only 4 of 16 streams) while keeping dense read access, fixing the cubic-cost and information bottlenecks that previously blocked scaling. On 18B and 28B MoE models it improves downstream scores by 4.0 and 3.1 points over prior mHC.

Cura 1T is a healthcare-specialized LLM trained through a human-supervised self-evolution loop, where a training agent iteratively plans capability improvements, evaluates against benchmarks, and refines the data mixture, rather than applying one-off medical data updates. It targets patient consultation, clinical reasoning over text/images, and interactive diagnosis with EHR tool use, reporting 0.940 on MedAgentBench and 0.796 pass@1 on AgentClinic.