Ant Group released Ling-3.0-flash-Fin, a finance-focused open-weights model. It scores 23 on the Intelligence Index and 24 on Finance & Accounting. The model matches MiniMax-M2.7’s Intelligence score with half the active parameters. It outperforms Ling-3.0-flash-VL in business knowledge accuracy but hallucinates more. It lags on difficult agentic tasks like AutomationBench-AA and Terminal-Bench. It uses more output tokens than both Ling-3.0-flash-VL and MiniMax-M2.7. It is text-only and available under an MIT license. It is accessible via OpenRouter with a rate-limited free endpoint.
What it unlocks
Using a finance-tuned model for research, valuation spreadsheets and report writing.
- 23 Intelligence Index score
- 24 Finance & Accounting Index score
- 5.1B active parameters per token
- 67k avg output tokens per task
What you need to act on it
- API access via OpenRouter
Sources