Model adaptation
Model adaptation is changing a model’s behavior or parameters in response to new data, context, feedback, or a target task or domain.
LLM Customization and Fine-Tuning
Modifying a pretrained model for a particular task, domain, or use case without training it from scratch.
- How it is used
- Positions adaptation as the middle ground between off-the-shelf models and training from scratch for enterprise tasks.
- How deep it goes
- Outlines the technique continuum and lifecycle, including evaluation gates, monitoring, retraining, and staged scaling deliverables.
- Examples
- Contrasts a hospital’s self-hosted clinical adapter with a support team’s prompting-and-retrieval baseline.
- What you can do after
- Evaluate which technique to start with, measure a baseline, and revisit the choice as the project develops.
Read how LLM Customization and Fine-Tuning explains model adaptation →
Context Engineering
Model customization is the adaptation of a general-purpose model to a particular task, domain, or organizational context.
- How it is used
- Positions customization as the bridge from a general-purpose model to a context-specific production solution.
- How deep it goes
- Contrasts intervention layers and operational burdens, including fine-tuning governance and rollback requirements.
- Examples
- Uses support tickets, legal documents, financial reports, and internal task demonstrations as potential enterprise training data.
- What you can do after
- Evaluate adaptation approaches by intervention layer and operational burden.
Also covered in
What to know first
- adapterAn adapter is a small trainable module added to a pretrained model to modify its behavior while leaving most of the model’s original parameters unchanged.
- foundation modelA foundation model is a broadly trained machine-learning model whose learned capabilities can be reused or adapted for a range of downstream tasks.
- fine-tuningFine-tuning is the continued training of a pretrained model on targeted data or objectives to adapt its parameters or behavior to a task, domain, or use.
- low-rank adaptationLow-rank adaptation (LoRA) is a parameter-efficient fine-tuning technique that adapts a model by learning low-rank updates to its weights, typically while keeping most original weights fixed.
Related concepts
- vendor lock-incontrasts with · Vendor lock-in is dependence on a provider, product, or technology that makes switching to an alternative difficult, costly, or impractical.
- adaptation techniqueuses · A particular method for adapting a base model, such as prompting, retrieval, or fine-tuning.
- foundation modeluses · A foundation model is a broadly trained machine-learning model whose learned capabilities can be reused or adapted for a range of downstream tasks.
- privacycontrasts with · Privacy is the protection of information about people or other sensitive subjects from inappropriate collection, access, use, or disclosure, including control over how it is handled.
- retrieval-augmented generationhas part · Retrieval-augmented generation is a technique that retrieves relevant information from external sources and supplies it as context to a generative model when producing an output.
Questions about model adaptation
What is model adaptation?
Model adaptation is changing a model’s behavior or parameters in response to new data, context, feedback, or a target task or domain.
Which Manning books explain model adaptation?
LLM Customization and Fine-Tuning and Context Engineering.
What should I know before learning model adaptation?
Manning books that teach model adaptation usually introduce adapter, foundation model, fine-tuning and low-rank adaptation first.
What is the difference between model adaptation and vendor lock-in?
Model adaptation is changing a model’s behavior or parameters in response to new data, context, feedback, or a target task or domain. Vendor lock-in is dependence on a provider, product, or technology that makes switching to an alternative difficult, costly, or impractical.
What are the parts of model adaptation?
Model adaptation includes retrieval-augmented generation.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.

