Ling 3.0 Flash

124B mixture-of-experts model with roughly 5.1B active parameters per token and a 262K context. A hybrid mode pairs the Ling line's speed with the Ring line's reasoning depth; hybrid-linear attention plus cluster-level hierarchical caching cut time-to-first-token on long inputs.

анонс 24.07.2026, веса открыты под MIT 07.08.2026. Производитель — лаборатория inclusionAI (Ant Group); в справочнике его нет

Understands

text

This model cannot see images or video — you can only give it text. That matters if you planned to show it screenshots or documents with pictures.

Produces

text, code

Contents verified with the vendor: 2026-08-18

Included in subscriptions

No catalogue entry marks this model as part of a plan yet. It will appear here once the plan contents are confirmed with the vendor.