Open-weight AI models from Chinese labs are reshaping how businesses deploy artificial intelligence, offering a path to lower costs, data privacy, and independence from vendor pricing changes. Unlike closed models such as GPT or Claude that you access through an API, open-weight models publish the trained weights, the core numerical parameters that encode what a model knows, allowing anyone to download and run them on their own hardware.

What Exactly Are Open Weights, and How Do They Differ From Other AI Models?

A large language model (LLM) is fundamentally a massive file of numbers. Those numbers are the weights: the learned parameters that encode everything the model knows how to do. Training produces them at enormous cost, but once they exist, running the model is comparatively cheap. Open weights means the lab publishes that file for anyone to download, run on their own hardware, fine-tune for their own purposes, and build products on top of it.

This differs from open-source software in the traditional sense. Open-source software includes the recipe: the training code, the data, and the full pipeline. Open-weight labs publish the finished weights and a technical report but keep the training data private. For practical purposes, the weights are what matter; you can use the model fully without the recipe.

It differs completely from closed models. GPT, Claude, and Gemini are accessed through an API or a chat window, with weights staying on the company's servers. You rent capability by the token, and the company can change pricing, deprecate versions, or cut you off. With open weights, the file is on your disk. It owes its maker nothing further, and neither do you.

Which Chinese AI Labs Are Leading the Open-Weight Movement?

Chinese AI laboratories have emerged as the primary drivers of open-weight releases. Moonshot AI announced Kimi K3 on July 16, with the full weights, all 2.8 trillion parameters of them, arriving as a public download by July 27. It will be the largest open-weight model ever released, and the release comes with the technical report describing architecture, training, and evaluations.

The two-stage launch illustrates how models work in practice. For the first eleven days, K3 works like any hosted model: you use it through Moonshot's products and pay by the token. On July 27, it becomes a folder on anyone's disk. Every question about pricing, access, capacity limits, and jurisdiction changes on that day.

Beyond Moonshot, other major Chinese labs publishing open-weight models include DeepSeek, Qwen, and GLM. These models are distributed primarily through Hugging Face, the main distribution hub where nearly every open-weight release lands within hours, alongside weights, documentation, and community discussion.

How to Get Started With Open-Weight Models

• Free Desktop Apps: Ollama and LM Studio are free desktop applications that make local running trivial. Install, pick a model from the catalog, and it downloads and runs on your machine with no account, API key, or data leaving your computer.
• Hosted APIs: Inference providers like OpenRouter, Together, Fireworks, Groq, and DeepInfra host open-weight models and sell API access. Because they all serve the same free file, margins compress toward zero and the savings pass through to users.
• Enterprise Platforms: Cloud platforms carry open-weight models inside enterprise procurement channels. AWS Bedrock, Azure AI Foundry, and Google Vertex all offer major open models with the compliance paperwork, billing integration, and security posture large organizations require.
• Lab Chat Interfaces: The labs themselves run free chat interfaces such as kimi.com, chat.deepseek.com, and chat.qwen.ai with no waitlist or invoice required.

What Are the Real Business Advantages of Open-Weight Models?

The immediate implication is price. Hosted open-weight models cost a fraction of the closed incumbents because a dozen providers compete to serve the same file. K3's own API launched at 30 cents per million cached input tokens. For a small firm, model spend drops from a budget line to a rounding error.

The deeper implication is ownership of the routine layer. A capable mid-size model on a single decent machine handles drafting, summarization, classification, translation, and first-line customer support. That covers most of what a small business actually asks of AI, with no per-seat licensing, no usage meter, and no monthly bill that scales with success.

For firms that handle confidential material, law, accounting, health, finance, local running means client data never leaves the premises. That single property settles the buy-versus-build question for a lot of professional services. Self-hosted weights run in whatever jurisdiction the enterprise chooses, allowing regulated industries that could never send data to a foreign API to run the same intelligence inside their own perimeter.

Open weights also eliminate vendor lock-in. Weights on your infrastructure survive the vendor's pricing changes, its deprecation schedule, its outages, and any diplomatic rupture between the country that made the model and the country that runs it. A downloaded model is a capital asset. An API subscription is a dependency.

Fine-tuning on proprietary data is another advantage. Open weights can be trained further on internal documents, code, and processes without any of that material leaving the building. The result is a model shaped to the business that the business fully controls. Past a certain volume, owning the serving stack is simply cheaper, and the crossover point keeps dropping as the models get more efficient.

What Are the Practical Limitations of Running Models Locally?

The full K3 is datacenter-scale. Practical deployment starts around 1.5 terabytes of memory, and Moonshot recommends supernode configurations of 64 or more accelerators. Almost nobody self-hosts a model that size, and almost nobody needs to. Within days of any major release, the community produces quantized versions, compressed variants that trade some capability for the ability to run on serious workstations, and the smaller siblings in each model family run acceptably on a recent laptop.

Latency is where self-hosting can be the better option. A model on your own hardware answers without a network round-trip, without rate limits, without queues, and without capacity walls. That last one is current: Moonshot has already stopped taking new API customers because demand outran its serving capacity. A hosted model can be full. A downloaded one never is.

The tradeoff runs the other way on raw speed for big models. A giant model squeezed onto modest hardware produces slow tokens, and specialized inference providers running custom silicon will beat your server rack on throughput. The practical answer is to right-size: match the smallest model that does the job to the hardware you have, and rent the big ones by the token when the task demands it.

The hybrid pattern is the sensible default. Evaluate through a hosted API, where switching costs are near zero. Self-host what proves out, where the compliance and cost advantages live. The same weights serve both stages.