Custom model
A custom model is a machine-learning model trained, adapted, or configured for a particular domain, task, or set of requirements.
Building Specialized AI Systems
A machine learning model tailored to a particular domain, task, or set of requirements rather than intended for broad general-purpose use.
- How it is used
- Targets mission-critical domain tasks requiring precision, consistency, deep expertise, or controlled handling of sensitive data.
- How deep it goes
- Explains fine-tuning as the primary specialization mechanism, including data curation, methods, evaluation, deployment, and decision trade-offs rather than neural internals.
- Examples
- Uses customer-support ticket classification, financial analysis, legal document classification, medical diagnostic support, code generation, and analyst-report summarization.
- What you can do after
- Evaluate/understand when specialization, prompting, RAG, or their combination best fits accuracy, knowledge, privacy, scale, and cost requirements.
Read how Building Specialized AI Systems explains custom model →
Rearchitecting LLMs
A specialized model is a machine-learning model adapted for a particular domain, dataset, or set of tasks and assigned as an expert within a larger model.
- How it is used
- Replaces a frequent, narrow agent call with a specialized small language model to address cost and latency.
- How deep it goes
- Details traffic filtering, task-specific pruning, recovery training, and a measured reference topology.
- Examples
- Uses structured tool-argument extraction and a confirmed table-booking request to illustrate specialization.
- What you can do after
- Evaluate a specialized model against a frontier API using held-out accuracy and shadow testing.
Data Analysis with LLMs
A language model trained or adapted for particular tasks or domains.
- How it is used
- Matches a language model to a specific data-analysis task, especially classifying reviews.
- How deep it goes
- Shows how to select and use specialist models and a fine-tuned model, without explaining training internals.
- Examples
- Uses review classification, a Hugging Face sentiment pipeline, and a fine-tuned model receiving review text alone.
- What you can do after
- Apply it via Hugging Face Transformers or the OpenAI client, then compare candidates on sample task data.
Understanding AI
A specialized model is a model adapted through additional training to the vocabulary, conventions, and reasoning patterns of a specific domain.
- How it is used
- Explains how general-purpose models become optimized for coding, medicine, law, or multimodal tasks.
- How deep it goes
- Traces specialization from a pretrained foundation through additional domain training, while retaining next-token prediction as the core mechanism.
- Examples
- Uses coding, legal, medical, image, audio, video, and document models as domain-specialization examples.
- What you can do after
- Evaluate how additional domain training changes a model’s learned behavior without changing its next-token prediction core.
Also covered in
- Fight Fraud with Machine LearningIn early access (MEAP)
- Data Without Labels
- Local Image GenerationIn early access (MEAP)
- Build Applications with Local AI Models on a Mac
- Designing AI SystemsIn early access (MEAP)
- Financial AI in Practice
- Generative AI in Action
What to know first
- 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.
- accuracyAccuracy is the degree to which a result agrees with a specified truth, reference, or criterion, commonly measured as the proportion of evaluated predictions that are correct.
- language modelA language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- promptA prompt is text or other input supplied to an AI model to provide instructions, context, or other information that guides its response or generated output.
Related concepts
- fine-tuningproduced by · Fine-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.
- generalist modelalternative to · A generalist model is a machine-learning model designed to perform a broad range of tasks or work across multiple domains rather than specialize in one narrow application.
- Modelfileproduced by · A Modelfile is Ollama’s declarative configuration file for specifying a model’s base, prompts, parameters, and other settings.
- Open WebUIuses · Open WebUI is an open-source web interface for interacting with large language models, including locally hosted models.
Questions about custom model
What is custom model?
A custom model is a machine-learning model trained, adapted, or configured for a particular domain, task, or set of requirements.
Which Manning books explain custom model?
Building Specialized AI Systems, Rearchitecting LLMs, Data Analysis with LLMs and Understanding AI.
What should I know before learning custom model?
Manning books that teach custom model usually introduce foundation model, accuracy, language model and prompt first.
What is the difference between custom model and generalist model?
A custom model is a machine-learning model trained, adapted, or configured for a particular domain, task, or set of requirements. A generalist model is a machine-learning model designed to perform a broad range of tasks or work across multiple domains rather than specialize in one narrow application.
What is custom model also called?
Custom model is also called specialized model or specialist model.
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.



