inclusionAI Releases Ling-3.0-tiny: An 8B Parameter MoE Model
-Cubie- · reddit · 2026-08-11
inclusionAI has open-sourced Ling-3.0-tiny, a much smaller version of Ling-3.0-flash, on Hugging Face. The model features 8B total parameters with only 1.3B active parameters.
Performance-wise, it falls between 4B and 12B Qwen and Gemma models. Thanks to the tiny active parameter count, it is expected to deliver massive tokens/sec inference speeds on most systems, highlighting the potential of tiny MoE architectures.
More from Models
- First Impressions: Gemini 4 Argon 'Incredible Writer' Beats Opus 5.5 and GPT 6 Astra — soumitrashukla9 · 2026-10-16
- Trying Grok bot surfaces an unpaid Cursor balance, blocking account merging — haydendevs · 2026-10-16
- TensorFold 1.0.12 runs 120B vision model on 1-2 DGX Sparks, adds Mac image input — HankYeomans · 2026-10-16
- Leaked screenshot shows unannounced Gemini 4 Argon gated to Ultra and paid Pro subscribers — il_-bro · 2026-10-16
- Google: Gemini 4 Argon lands in Antigravity, code submissions up 35%, rollbacks down ~33% — tekbog · 2026-10-16
- Google's "ungrounded beliefs" classifier injections derail Opus 4.1's reasoning chain — liminal_bardo · 2026-10-16