Glossary
Open-weights model
In short
A model whose trained weights are released publicly, so it can be run and hosted by anyone, not just its original lab.
Closed models are only available through their maker’s API. Open-weights models release the trained weights, so the wider ecosystem can run, host, and serve them, which tends to push prices down and choice up.
Several strong models are open-weights, and they often deliver excellent quality for the price, which is why they show up so often in cost-conscious setups.
In LLMWeave
Several of the models in LLMWeave are open-weights, including some of the free ones, which is part of how a capable free tier is possible.
Related terms
Mixture of experts
A model architecture that activates only a relevant subset of its parameters for each input, giving large-model capability at lower running cost.
Multi-model AI
Using more than one AI model on the same task, instead of relying on a single model for every answer.
Model routing
Sending each task, or each part of a task, to the model best suited for it, rather than using one model for everything.
Try multi-model on your task
One prompt, several models, one answer. Free to start, no card.
Get started