Open-weight model
Also known as: open-weights model, open weight model, open-weight LLM
An open-weight model is an AI model whose trained parameters, or weights, are released publicly so anyone can download, run, and often fine-tune it on their own hardware.
That differs from closed models available only through a company's API. While open-weight models are often called open source, many don't share their training data or code, and their licences may restrict commercial use, so 'open' covers a wide range.
Open-weight model, explained at three levels
At home
Most AI chatbots run on a company's servers, and you can only use them through their app. An open-weight model is one you can download and run on your own computer.
At work
Open-weight models let organizations run AI on their own infrastructure, giving more control over data, cost, and customization.
At home
The 'weights' are the billions of numbers the model learned during training. Publishing them is like sharing a finished recipe, even if the company keeps its shopping list and kitchen notes private.
At work
Licences vary widely: some are permissive (Apache 2.0, MIT), while others limit commercial use, user counts, or certain applications. Teams self-host using tools like vLLM or llama.cpp, or use cloud providers that host open models.
At home
Once weights are public, anyone can remove safety features. In my opinion, that's a real risk, but it's balanced by the benefit of researchers and smaller companies not depending on a few giants.
At work
Evaluation should cover licence terms, model provenance and country of origin, built-in censorship or bias, security of downloaded files, ongoing updates, and total cost of self-hosting. Open weights support data sovereignty and air-gapped deployments, which matter in government and regulated sectors.
Real-world examples
At home
Thomas downloaded an open-weight model and ran it on his gaming PC. He could chat with it offline and knew none of his conversations were sent to a company's server.
At work
Lena is an IT director for a Canadian municipality. She chose an open-weight model hosted in a Canadian data centre so residents' data stayed in the country and under the city's control.
How it compares
| feature | Open-weight model | Closed model |
|---|---|---|
| Weights available | Yes, downloadable | No, API only |
| Training data shared | Usually not | No |
| Self-hosting | Possible | Not possible |
| Top capability | Close behind leaders | Often the leaders |
Open-weight models give control. Closed models often lead on raw capability.
For analysts
Before adopting an open-weight model, I'd have someone read the licence carefully. 'Open' can still mean usage limits that matter for a commercial product.
History & origin
Meta released its LLaMA model to researchers in February 2023, and the weights leaked publicly within about a week. Meta then released Llama 2 in July 2023 under a licence allowing most commercial use but restricting very large companies. France's Mistral AI released Mistral 7B under the permissive Apache 2.0 licence in September 2023. Toronto-based Cohere published weights for its Command R models for non-commercial use in 2024. The Open Source Initiative published its Open Source AI Definition in October 2024, setting a higher bar that most open-weight models didn't meet. DeepSeek's R1, released in January 2025 under the MIT licence, and Alibaba's Qwen models intensified competition, and OpenAI released its first open-weight models since GPT-2, gpt-oss, in August 2025. In my view, open-weight models are the most important check on a handful of companies controlling AI.