AI Foundationspredict · compress · act
act XI

The Open-Weight World

By mid-2025 China had released more public models than the rest of the world combined. Here is who they are, and why they chose to give them away.

46

The Chinese Labs

frontieras of 2026-09
before this →DeepSeek

DeepSeek looked like a surprise, but it was one move inside a much larger one. Chinese labs had been releasing weights steadily for years, and by July 2025 Chinese state media counted 1,509 of the world’s roughly 3,755 public models as Chinese — more than any other country. That count is from mid-2025 and the cadence has not slowed since. The strategy is deliberate, and it is not only about generosity.

LABMODEL FAMILYLICENCEAlibabaZhipu AIMoonshotMiniMaxShanghai AI LabTencentByteDanceBaiduQwen (text · vision · audio · code · many sizes)GLM (chat · agent · multimodal)Kimi (long context · large MoE)MiniMax (MoE · audio · video)InternLMHunyuanDoubao / SeedErniemostly Apache 2.0permissivemodified MITpermissivevariesvariesopen-weight linespivoted to open
the main Chinese model families. Colour marks how permissive the licence usually is — always check the model card, terms change.

The families

Qwen (Alibaba). The broadest line by far: dense and mixture-of-experts models, sizes from tiny to huge, and text, vision, audio, and code variants under one naming scheme. It has become the default base for fine-tuning worldwide, and most newer open models are measured against it. The line has kept moving: the 3.5 generation was followed by 3.6 and then Qwen3.8, and in 2026 Alibaba opened the weights of a Qwen-Max-class model — 2.4 trillion parameters — for the first time. That is a different order of release from putting a mid-size model out for free.

GLM (Zhipu AI, also written Z.ai). Strong general models with an emphasis on agents and tool use, released on a steady cadence.

Kimi (Moonshot AI). Known first for very long context, then for Kimi K2, a large mixture-of-experts model released with open weights and tuned for agentic work.

MiniMax. Efficient mixture-of-experts models with a focus on long context, plus audio and video lines. InternLM (Shanghai AI Lab), Hunyuan (Tencent), Doubao / Seed (ByteDance), and Ernie (Baidu, which had pursued a closed strategy before pivoting) round out the list. Several of them now open their weights as a matter of course.

Why give it away?

Four reasons, none of them charity

the same motives as anywhere — with one extra

Distribution. A free download is the marketing: it earns users, developers, and citations. Ecosystem. Whoever’s model everyone fine-tunes on sets the standard others must follow. Talent. Strong open work attracts researchers. The squeeze. American export controls limited access to the best chips, so efficiency and openness became a way to compete without the biggest clusters — an idea we pick up in the next chapters.

a real differenceMany Chinese labs ship under Apache 2.0 or MIT, while several Western labs use custom community licences with usage limits. That difference, not the model quality, is what made these weights adoptable.
RELEASEweights go public
ADOPTIONthe world fine-tunes on your base
STANDARDyour choices become everyone's defaults

What to watch

A model family is not one model but a lineage — sizes, versions, and modalities that share training choices and a tokenizer. When a family ships under a permissive licence, its tokenizer and chat format quietly become a standard, and everything downstream inherits both its strengths and its quirks. The next chapter widens the view beyond China, and then we look at the constraint that shaped all of it: compute.

introduces →model familyQwenGLMKimiMiniMaxInternLMHunyuan
← previousDeepSeeknext →The Other Open Lineages