Ant Group releases Ling-3.0-flash-Fin, an open-weights AI model for finance
Artificial Analysis reports that the model matches MiniMax-M2.7’s Intelligence Index score of 23 with roughly half the active parameters per token: 5.1 billion versus 10 billion.
TLDR
Artificial Analysis reports that Ant Group released Ling-3.0-flash-Fin, a text-only, open-weights model built on Ling-3.0-flash. Ant Group says it developed the model with financial institutions and industry experts to support financial research, including checking sources, building valuation spreadsheets and writing reports.
Artificial Analysis gives it 24 on its Finance & Accounting Index, matching Ling-3.0-flash-VL. It reports higher business knowledge accuracy for Fin than VL—17% versus 11%—but also higher business knowledge hallucination: 33% versus 19%.
The evaluator notes that Fin uses 5.1 billion active parameters per token but has 124 billion total parameters.
Combined views
1.9K
1 Source, first seen 14d ago