inclusionAI/Ling-3.1-flash · Hugging Face
retained text
With **560B** total parameters, **25B** active parameters per token, and a **1M-token context window**, wait what? 3.0 was just 127B
document fetched 2026-10-15 00:46 · canonical https://reddit.com/r/LocalLLaMA/comments/1x5tsni/inclusionailing31flash_hugging_face