Glossary

AI Systems

Computer programs designed to perform tasks that typically require human intelligence, such as learning, reasoning, and problem-solving. They're like smart machines that can think and act on their own.

AI21 Labore

AI21 Labs is an Israeli company specializing in NLP and the development of advanced AI systems that can understand and generate natural language. It provides foundation models and AI solutions tailored for enterprise use.

API

Application Programming Interface is a set of rules, protocols, and tools that enable different software components to communicate and integrate AI capabilities. It's like a bridge that connects AI models to other applications, allowing them to share data and functions seamlessly.

Amazonas-Felsen

Fully managed service offered by AWS that enables developers to build and deploy generative AI applications using a variety of foundation models (FMs) and LLMs from leading AI companies. It simplifies the integration of AI capabilities into applications, providing a single API for accessing multiple models.

Anthropisch

Company and research organization focused on developing and applying artificial intelligence in a safe and beneficial manner. It's like a guardian of AI, ensuring that these powerful technologies are used responsibly.

Chunk

A chunk in AI is a small, manageable piece of data or information that is processed or analyzed as a unit.

Cohere

Canadian multinational technology company specializing in artificial intelligence, particularly in LLMs and natural language processing for enterprise applications. It provides a platform for businesses to deploy generative AI solutions securely and efficiently.

Context

Refers to the surrounding information or environment that provides meaning and relevance to the input or output of an AI model. It's like the backdrop that helps the AI understand the nuances and intentions behind the text or data it is processing.

Fine-Tuned Model

Pre-trained model that has been further trained on a smaller, task-specific dataset to adapt it to a particular use case or domain. This process enhances the model's performance and relevance for the targeted task.

Frequency Penalty

Parameter used in LLM to control the likelihood of word (tokens) repetition in general context of the generated text. It discourages the model from using the same words or phrases repeatedly, promoting more diverse and varied output.

Function Calling

Function calling is when a program invokes a predefined function to perform a specific task.

GPT (Generative Pre-trained Transformer)

Advanced AI model known for generating text that's almost indistinguishable from human writing. It’s like a robotic author with a knack for mimicking human conversation and composition.

Gemini

Collection of multimodal LLMs developed by Google, designed to understand and process various types of data including text, images, audio, video, and code. It is the successor to Google's Bard and represents a significant advancement in Google's generative AI capabilities.

Google KI

Artificial intelligence research and development division of Google, focused on advancing AI technologies and integrating them into various Google products and services. It's like the brain behind many of Google's innovative features, driving advancements in machine learning, natural language processing, and more.

Hackathon

Event where participants, often including prompt engineers, developers, designers, and other specialists, collaborate intensively over a short period, typically 24 to 48 hours, to develop new software, applications, or features. It's a marathon of innovation aimed at solving specific problems or creating new products.

Hugging Face

Popular open-source platform for natural language processing (NLP) and machine learning. It provides pre-trained models, tools, and a community to help developers build and deploy AI models.

Instruction

Specific directive given to an AI within a prompt. It's like giving your AI assistant a clear, step-by-step recipe to follow, ensuring it knows exactly what you want it to do.

JAIS

JAIS is an open-source large language model developed in the United Arab Emirates, specifically designed to understand and generate text in both Arabic and English. It is named after Jebel Jais, the highest mountain in the UAE, and is intended to bring the benefits of generative AI to the Arabic-speaking world.

Latency

Refers to the delay or time it takes for a system or process to respond to a request or input. In the context of AI, it's the time between when a command is given and when the system reacts.

Log

Record of events, activities, or data points that are generated by an AI system or its components. It's like a digital diary that tracks what the system is doing, helping in debugging, monitoring, and optimizing its performance.

Microsoft Azure

Cloud computing platform that offers a wide range of services, including artificial intelligence, to help developers and organizations build, deploy, and manage intelligent applications. It's like a comprehensive toolkit for creating and scaling AI solutions in the cloud.

Mistral AI

French artificial intelligence startup that specializes in developing and providing large language models and generative AI solutions. It offers both open-source and commercial models, known for their efficiency, customization, and multilingual capabilities.

OpenAI

Research organization and company focused on developing and directing artificial intelligence in ways that benefit humanity. It's known for creating advanced AI models like ChatGPT and DALL-E.

Performance Optimization

Process of improving an AI model's efficiency and effectiveness. It's all about making your AI assistant smarter, faster, and more accurate over time. It's like fine-tuning a machine to run smoothly and deliver better results.

Perplexity

AI-powered search engine and conversational research tool designed to provide accurate, comprehensive, and verifiable answers to user queries. It acts as a research partner, summarizing information from the internet and citing trusted sources.

Presence Penalty

Parameter used in generative AI models to control the repetition of tokens, words or phrases in the generated text. It discourages the model from using the same elements multiple times, promoting diversity and novelty in the output.

Prompt

A prompt is the input given to an AI system, guiding its response. It's like a conversation starter or task instruction for your AI assistant, setting the stage for interaction.

Prompt Technik

Prompt engineering is the art and science of crafting effective prompts for AI systems. It's like being a skilled conductor, guiding the AI orchestra to produce the perfect symphony of responses.

Query

A query is a request for information or data, often used in databases and AI systems to retrieve specific answers or results.

RAG (Retrieval-Augmented Generation)

RAG, or Retrieval-Augmented Generation, is a technique that enhances the accuracy and reliability of generative AI models by fetching facts from external sources. It's like having a research assistant that ensures the AI's responses are grounded in up-to-date, verified information.

Reasoning

Ability of a system to draw inferences, make decisions, and solve problems based on the information it has been given. It is a fundamental aspect of AI that enables machines to think and act in ways that mimic human intelligence.

Role Specification

Role specification is the practice of assigning a specific persona or expertise to an AI during interaction. It's like asking the AI to put on a virtual hat, transforming it into a subject matter expert for more tailored responses.

Search Engine

AI Software system that helps you find information on the internet by searching through a vast index of web pages.

Semantic Search

Type of search that understands the context and intent behind a search query, providing more relevant and accurate results.

Serverless

Cloud computing execution model where the cloud provider manages the infrastructure, allocating machine resources on demand, and the user is not concerned with server management. It's a paradigm that allows developers to focus solely on writing and deploying code without worrying about the underlying servers.

Task

In AI context, a task is a specific action or objective you want the AI to accomplish. It's like giving your digital assistant a to-do list item, guiding its efforts towards a particular goal.

Temperature

In the context of machine learning and deep learning, "temperature" is a parameter used to control the randomness of the output in models, particularly in those involving probabilistic predictions.

Tokens

Tokens are the smallest units of text or data that a computer can process. They are like the building blocks of language, helping AI models understand and analyze text.

Vector Embeddings

Numerical representations of words or data points in a high-dimensional space, allowing for semantic comparisons and analyses.

Vision

Ability of machines to interpret and understand visual data from images and videos.

Web-Scraping

Web scraping is the process of automatically extracting data from websites, often using AI and machine learning to handle dynamic content and anti-scraping measures. It's like a digital librarian that gathers information from the internet for you.

No results found

We're continually expanding our artificial intelligence glossary. If there are concepts you think we're missing, let us know by emailing us at hello@promptitude.io, and we'll add them soon.

SIGNUP

Sind Sie bereit, loszulegen?

GPT Prompts für

Technische Redakteure Translation & Localization Managers
Erfolgsgeschichten Wall of Love

Anwendungsfälle

Benutzer-Onboarding

KI-Textgenerierung

GPT-Hackathons

Kreative Unterstützung

Automatisierung von Sitzungsprotokollen

Interaktive Leadmagnete

Produkt

Preise

Modelle - Integrationen

Funktionen - Dienstleistungen

Roadmap - Changelog

Ressourcen

Blog Glossary Hilfe-Center Wie es funktioniert ChatGPT-Alternative Copilot Alternative

Die KI-Landschaft entwickelt sich mit Lichtgeschwindigkeit. Machen Sie Ihr Unternehmen mit Promptitude zukunftssicher und verwalten Sie alle Ihre Prompts und KI-Anbieter an einem Ort, teilen Sie sie mit Ihrem Team und integrieren Sie KI in Ihr Unternehmen.

Zertifiziert als innovatives Unternehmen durch das deutsche Bundesministerium für Bildung und Forschung.

Nutzungsbedingungen - Datenschutz

Zugang zu kostenlosem PDF erhalten

Entfalten Sie die Kraft von Prompt Engineering 101

Demystifying AI: Your Go-To Glossary