inclusionAI · MoE Architecture
Ling-2.6-flash
Open-source 104B-parameter MoE instruct model with 7.4B active params, tuned for agent workloads. Hybrid linear architecture hits ~340 tokens/s and is engineered for token-efficient, production-scale automated tasks.
Open the source ↗Similar Language Models releases
- MiniMind
jingyaogong · Train An LLM Cheaply · Week 37, 2026
- Inkling-Small
Thinking Machines Lab · MoE Architecture · Week 31, 2026
- Instella-MoE
AMD · MoE Architecture · Week 31, 2026
- Spark-X2.6-1.7B
XHToken · On-device LLM · Week 45, 2026
- Clover-35B-A3B-Preview
DiffuSpace · Diffusion MoE LLM · Week 45, 2026
- Groot-280B-A16B-261023
GrootTeam · Open reasoning MoE · Week 45, 2026
More from inclusionAI
- Ming-Image-0.1-Design
inclusionAI · Design-Focused Image Model · Week 40, 2026
- LLaDA-Image
inclusionAI · Diffusion Image Model · Week 36, 2026
- Ming-Lite-Omni
inclusionAI · Multilingual · Week 36, 2025
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.