Ling-3.0-flash, an open-weight model from Ant Group's InclusionAI, is now live on AI/ML API (@aimlapi).
• the model delivers some serious agentic capability with its 124B total param size (only 5.1B active) • Hybrid attention makes its 256K context cheaper to process. • It was trained on full agent trajectories, including tool recovery. • The payoff is lower cost across long, multi-step agent runs.
AI/ML API is actually keeping it for free until August 6: aimlapi[.]com
Ling-3.0-flash is live on AI/ML API 💎 Created by @AntLingAGI , Ling-3.0-flash is a hybrid-reasoning MoE built for production-scale agents, delivering flagship-...