SIG-2026-004 · Open Chinese models
In brief: According to Hugging Face, open-weight Chinese models took 41% of downloads versus 36.5% for American ones; about 80% of US AI startups use Chinese models somewhere.
Date / period: 2026.
Geography: China; global.
Type and stage: Research or data.
Scenario: DS — Decentralized — Synthetic.
Why this scenario: The person chooses and assembles the tools and is responsible for the outcome.
Collection trend: AI as the go-between for people and the world — AI moves into the physical world and becomes an agent that searches, buys and decides for us. The question is whose agent it is and what values it was trained on.
Macrotrend drivers: MT-047 — Advanced AI systems and cognitive computing; MT-002 — Human–machine interaction; MT-081 — Ecosystems, super-apps and the platform economy.
Life domains: Everyday life, consumption and leisure; Finance.
Thematic tags: ai agents
Description
According to Hugging Face, open-weight Chinese models took 41% of downloads versus 36.5% for American ones; about 80% of US AI startups use Chinese models somewhere. They can be downloaded, run locally and fine-tuned.
Why this is a signal
Your own AI without a subscription or someone else’s cloud becomes available to all.
Where the signal comes from
Research or data — tech-insider.org.
Limits of the conclusion
Figures come from the study or survey given; sample and method are the authors’. Assigning it to a Decoder world is our reading.
Links
- This signal in Russian.
- Other signals of this trend: CS · Agents pay for us; CS · AI advisers for collectors; CE · The “AI+” plan.
Source: tech-insider.org.
All signals in this collection: Signals: big trends, November 2026