PulseAugur
EN
LIVE 10:38:55

Chinese LLMs Dominate Top 300 Open Model Downloads

By PulseAugur Editorial · [1 sources] ·

A new analysis of open-source Large Language Models (LLMs) reveals that 66% of the top 300 models by recent downloads originate from China. The study, which focuses on trailing 30-day downloads on Hugging Face rather than cumulative figures, attributes models based on their training origin rather than the uploader's organization. This method highlights that a significant portion of popular LLMs, particularly in the GGUF format used for local execution, are Chinese-developed. AI

IMPACT Highlights the significant global reach and adoption of Chinese-developed LLMs, particularly for local deployment.

RANK_REASON Analysis of open-source LLM download data [lever_c_demoted from research: ic=1 ai=1.0]

Read on dev.to — LLM tag →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Chinese LLMs Dominate Top 300 Open Model Downloads

How we ranked this

Signal score
44 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Analysis of open-source LLM download data [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, other
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. dev.to — LLM tag TIER_1 English(EN) · AI OpenFree ·

    We ranked the top 300 open LLMs by real downloads. 66% of them are Chinese.

    <h1> We ranked the top 300 open LLMs by real downloads. 66% of them are Chinese. </h1> <p>Every "top open model" list I could find ranks by cumulative downloads. Cumulative rewards age: a<br /> model from 2022 outranks one that is actually being used today. So we built the other …

RELATED ENTITIES

RELATED TOPICS