A next-generation general-purpose large language model for AI-native agents, built on a core design philosophy of high intelligence density x high-value tokens, focused on task execution rather than pure conversation.

Intelligent Q&A and content understanding for health insurance scenarios, supporting analysis and interpretation of medical records, reports, policies, and clauses, key-point extraction, and multi-turn dialogue.

Transcribe audio to text quickly and accurately for multiple scenarios.

Text-to-speech with natural, expressive speech in multiple voices.

All-in-one TTS and voice cloning: clone in seconds with high fidelity.

Delivers four core parsing capabilities: intelligent document classification, precise ID and certificate recognition, faithful layout restoration, and smart key information extraction.

Built for medical scenarios, combining document classification, layout parsing, and professional information extraction in one solution.