Hugging Face Blog·2026-08-14 08:00· 2026-08-14

State of Open Models: Summer 2026 Observations

AI summary

Hugging Face published its open-source model observatory report for January–August 2026. Hub data shows Chinese labs released the largest open-source models by parameter count in most months, with Chinese monthly peaks between 754 billion and 2.78 trillion parameters, while US models stayed below 130 billion in five of seven months.

Selection record

Threshold 60Official, first-handFirst 62Second 62

AdmittedSum of both 124 ≥ twice the threshold 120

Source tier: Official, first-hand; this tier's threshold is 60
Pre-filter: passed：Hugging Face开源模型生态报告，AI技术相关
Why it was chosen: Seven months of Hub data break down the open-source landscape, with differences between downloads and likes, Qwen's niche and agent traffic changes offering a transferable analytical framework.

A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.

Source: Hugging Face Blog · huggingface.co

#Research #Open-source ecosystem #Agents #Data and training #Hugging Face