Compute
Compute is the processing capacity, resources, or computational work involved in executing software and other computational workloads.
Machines that Think
Processing capacity available to run computational workloads.
- How it is used
- Frames compute as the capacity needed to train and run AI models and as a constraint on frontier development.
- How deep it goes
- Explains parallel hardware and supply constraints at an overview level, including flexibility and portability trade-offs.
- Examples
- Contrasts GPUs with TPUs and AWS accelerators, and describes cloud partnerships that secure compute capacity.
- What you can do after
- Evaluate and understand how compute availability and hardware choices constrain AI development and deployment.
Timeless Algorithms, The Foundational Ideas
A computational budget is the fixed amount of computation available for a task.
- How it is used
- Constrains neural-network training and deployment, and frames how model size and training data are allocated.
- How deep it goes
- Defines training compute, gives its approximate parameter-token relationship, and explains balanced allocation without specifying an exact budget formula.
- Examples
- Contrasts total training work with real-time prediction in an autonomous vehicle and illustrates tradeoffs between parameters and training tokens.
- What you can do after
- Apply the compute-efficient frontier to choose model capacity and training-data volume for a fixed computational budget.
Read how Timeless Algorithms, The Foundational Ideas explains compute →
Introduction to Generative AI, Second Edition
Compute is the computational processing capacity required to train, run, or adapt machine-learning models.
- How it is used
- Connects compute capacity to competition over AI infrastructure, strategic autonomy, access, and investment.
- How deep it goes
- Links compute to infrastructure and operating costs, but does not explain how processing capacity works technically.
- Examples
- Cites AI providers whose inference and infrastructure expenses far exceed subscription revenue.
- What you can do after
- Evaluate how compute and infrastructure costs shape AI investment and potential affordability.
Read how Introduction to Generative AI, Second Edition explains compute →
Inside AI
A computational resource is processing, memory, storage, or related capacity consumed by a computational process.
- How it is used
- Frames computing power as a constraint on AI progress and an environmental cost of training and running models.
- How deep it goes
- Connects compute demand to chip-density and heat limits, then outlines efficiency and data-reduction responses without implementation detail.
- Examples
- Cites chess victories, a reported 300,000-fold rise in AI computing power, and emissions from building and training an AI system.
- What you can do after
- Evaluate AI resource costs and prioritize efficient algorithms and reduced training-data, storage, and transmission demands.
Also covered in
- Distributed Machine Learning Patterns
- Azure Security
- Designing Deep Learning Systems: A guide for software engineers
- Understanding AIIn early access (MEAP)
- GitHub Actions in Action: Continuous integration and delivery for DevOps
- ScyllaDB in Action
- The Complete Obsolete Guide to Generative AI
- Cloud Observability in Action
- How Large Language Models Work
- Building LLM Applications with DSPy
- Grokking Concurrency
- Snowflake Data Engineering
What to know first
- language modelA language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- memoryMemory is a storage capability or mechanism for retaining information or computational state for later access or use.
- artificial intelligenceArtificial intelligence is the field of computing concerned with creating systems that perform tasks associated with intelligence, such as learning, reasoning, perception, and decision-making.
- containerA software container is an isolated execution environment that packages an application and its dependencies, typically sharing the host operating-system kernel.
Related concepts
- Moore's lawmeasured by · Moore’s law is the empirical observation that the number of transistors on an integrated circuit historically tended to double approximately every two years.
- neural networkused by · A neural network is a computational model composed of interconnected, parameterized units or transformations that learn to map inputs to outputs.
- reasoninguses · Reasoning is the process of deriving conclusions, solutions, or actions from available information through inferential steps.
- infrastructureused by · Infrastructure is the foundational physical, digital, and operational resources and systems that support the operation of organizations, computing systems, or services.
- model training serviceused by · A service that provides computation resources and job management for executing model-training code, producing trained models, and tracking training execution.
Questions about compute
What is compute?
Compute is the processing capacity, resources, or computational work involved in executing software and other computational workloads.
Which Manning books explain compute?
Machines that Think, Timeless Algorithms, The Foundational Ideas, Introduction to Generative AI, Second Edition and Inside AI.
What should I know before learning compute?
Manning books that teach compute usually introduce language model, memory, artificial intelligence and container first.
What is the difference between compute and mind?
Compute is the processing capacity, resources, or computational work involved in executing software and other computational workloads. The mind is the set of cognitive and experiential faculties associated with thought, perception, intention, and understanding.
What is compute also called?
Compute is also called computing power, computational resources, processing power, computational power or computing resource.
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.



