Ling 3 0 Flash Vl Release

1 update · 1 outlet · 8 Sep 2026

Timeline

1. 8 Sep 2026Inclusion AI releases Ling-3.0-flash-VL, a 124B multimodal MoE with 1M context

   Ling-3.0-flash-VL shows the trend of packing vision, video, and million-token context into sparse MoE models small enough to run with only 5.5B active parameters per token.

   1 source · thin-sourcing, excerpt-only

All sources (1)

1. r/LocalLLaMA2026-09-08

Related subjects

• Deepseek Flash 4 1 BetaModels
• Nex N2-5 Mini $35B ReleaseModels
• Qwen Drive 1 0 $4B ReleaseModels