Model size

Knowledge graph concept · also referred to as size, mid-sized, smaller, size of weights, mid-sized models, 27b, 20gb, active params, 18b active params, 13b, small, small enough, model sizes, parameter count, bigger Pro model

Mentioned in

• The barbell effect will deprioritize mid-sized models X thread · 2026-09-18
• Qwen3.8 27B compressed to 5.95 GB, running in a browser X thread · 2026-09-18
• Where DeepSeek V4.1 Pro could land X thread · 2026-09-16
• I trained an AI model on my phone X thread · 2026-09-13
• DeepSeek V4.1 Flash is built for datacenters X thread · 2026-09-10
• The ideal model could run locally X thread · 2026-08-14
• Local AI will favor LPDDR and MoE X thread · 2026-08-10
• Report Per-Session and Aggregate Token Throughput X thread · 2026-07-21
• Popularity rankings for open models X thread · 2026-06-20

Related concepts

• Inference throughput 4
• Open-weight language model 4
• AI inference pricing 3
• KV cache 3
• Mixture-of-experts model 3
• AI benchmark 2
• AI model competition 2
• Local model deployment 2