AI Glossary Term
What does “Mid-training” mean?
An intermediate training phase between pretraining and post-training, often used for capability or domain adjustments.
How Mid-training works
Example
After broad pretraining, a lab continues training on a higher-quality mix with more code and math, and on longer documents to extend the context length, before the final instruction-tuning stage.
Why it matters
It lets developers strengthen particular skills without repeating the full pretraining. The term is used loosely, and boundaries between stages vary by lab.
Common questions
Is it a standard term?
It is informal and used differently across labs.
Why add this stage?
To improve specific capabilities or extend context length at lower cost than starting over.
How is it different from fine-tuning?
It is usually larger scale and still focused on general capabilities, while fine-tuning typically targets a specific task or behavior.