Inference
Running an already trained model to produce output. It differs from training; its cost depends on the model and processed volume.
#A practical artificial intelligence glossary
Short, precise definitions to understand product claims, ask better questions and evaluate tools. 48 terms organized by use.
48 terms
Foundations
Model card
A document describing a model, intended uses, limitations and information about training or evaluation. Check which version it covers.
#AI model
A system trained on data to produce an output from an input. One model can be offered through several applications.
#Large language model (LLM)
A model trained to process and generate language. It can write or summarize, but it is not a guaranteed database of facts.
#Token
A unit of text processed by a model: a word, part of a word or punctuation. Limits and costs are often expressed in tokens.
#Context window
The maximum amount of information a model can consider in one request, including input and output.
#Interacting with a model
Structured output
A response following a defined format, such as a JSON schema. Valid formatting does not prove the values are correct.
#Prompt injection
An attempt to make a model follow instructions inserted into input or documents instead of the authorized task. Treat external documents as untrusted data.
#Prompt
Instructions and context given to an AI system. A useful prompt states the intended result, constraints and format.
#System instruction
An instruction shaping an assistant’s behavior. It does not replace technical access and data controls.
#Multimodal
The ability to process or produce several content types, such as text, images, audio and video.
#Temperature
A parameter affecting output diversity in some models. A lower value does not guarantee accuracy.
#Answers and sources
Optical character recognition (OCR)
Converting characters in an image into text. Recognition does not validate reading order, table structure or value accuracy.
#Reranking
Reordering retrieved documents by relevance to a query. It does not replace checking the content of the documents.
#Provenance
The origin and history of information: document, author, date and transformations. Provenance helps trace the support for a claim.
#Source freshness
How well a source’s date fits the question. Distinguish publication date, event date and the corpus’s last synchronization.
#Hallucination
A plausible but false, invented or unsupported answer. A source or citation can also be fabricated.
#Grounding
A method linking an answer to identifiable documents. It helps verification, but a source can still be misread.
#Retrieval-augmented generation (RAG)
An architecture retrieving passages from a corpus before passing them to a model for its answer. Quality also depends on the documents and retrieval.
#Source citation
A reference shown with an answer so a claim can be traced. Open the source and check the passage, date and context.
#Architecture
Embedding
A numerical representation used to compare semantic similarity. Nearby items are not necessarily equivalent or true.
#Vector database
A system storing vectors and retrieving nearby items. In RAG it helps select passages but does not verify their accuracy.
#Fine-tuning
Additional training on targeted examples to change model behavior. It does not replace an up-to-date document collection.
#Application programming interface (API)
A technical interface through which an application requests an AI service. Costs, quotas and data rules depend on the plan.
#Agents and automation
Idempotency
A property where repeating an operation has no additional effect compared with executing it once. Useful for preventing duplicates during retries.
#Least privilege
Granting only permissions needed for a task and for the required duration. An agent should not receive all access by default.
#Retry
Re-execution after failure. Limit attempts, distinguish temporary from permanent errors and prevent duplicate writes.
#AI agent
A system combining a model, goals, tools and decision steps to complete a task. Its autonomy should be bounded by permissions.
#Tool calling
A mechanism through which a model asks an application to run a function such as search or calculation. The application controls execution.
#Workflow
A sequence of steps connecting people, data and tools. Reliable automation plans for errors, recovery and approval.
#Human review
A decision or check performed by a person before or after a system action, useful when errors could have significant consequences.
#Evaluation
Critical error
An error whose consequence makes an output unacceptable even if other criteria are satisfied. Define it before the trial, such as a wrong recipient, incorrect amount or ignored permission.
#Abstention
Declining to answer or act when information or authorisation is insufficient. Evaluate it on explicitly out-of-scope cases.
#Acceptance criteria
Observable conditions set before a trial to determine whether a result is usable. Include errors that require rejection.
#Cost per accepted result
Total trial cost divided by results passing checks, including service, calls, preparation and correction. It differs from generation price.
#Observability
Understanding system behaviour through logs, metrics and traces. Retain useful evidence without unnecessarily logging sensitive data.
#Evaluation set
A stable set of representative cases and criteria used to compare versions of an AI system on a defined task.
#Benchmark
A standardized evaluation comparing systems under a stated method. A strong public score does not guarantee quality on your data.
#Regression
A decline in behavior that previously worked, sometimes after a change to a model, prompt or data.
#Latency
The time between a request and a usable result. Also measure retrieval, tool execution and human review time.
#Data and rights
Data portability
The technical ability to export and reuse data in another system. Check formats, attachments, metadata and reimport; downloading alone does not prove portability.
#Reversibility
The ability to leave a tool or return to an earlier state with controlled effort. Verify it with a restoration trial rather than an export promise alone.
#Self-hosting
Operating software on infrastructure you manage. It requires maintenance and backups and does not guarantee all calls remain local.
#Data residency
Locations where data is stored or processed. The publisher’s country is insufficient: check regions, subprocessors and contractual documents.
#Personal data
Information that identifies a person directly or indirectly. Check necessity and processing terms before sharing it.
#Data retention
The period for which a service stores inputs, outputs or logs. It varies by product, plan and settings.
#Open weights
A model whose trained parameters are distributed under a license. It does not always mean the data or code is open.
#Usage license
Terms governing use, redistribution or modification of a model, tool or content. Check the exact license before commercial use.
#No term matches this search.