Ant launches Ling-3.0-flash-Fin: 124B MoE, 5.1B active, 262K context

Article automatically generated from technical news.

Ant's Ling team has released Ling-3.0-flash-Fin, a finance-focused mixture-of-experts model built on Ling-3.0-flash. The model has 124B total parameters and about 5.1B active. OpenRouter lists a 262,144-token context window, up to 32,768 output tokens and support for tools and tool_choice. The release targets financial information retrieval, research, valuation modeling, report preparation, long reports and complex workbooks. The official evaluation set includes FinFIRST, Fin