Glossary

Agentic AI pursues complex goals with autonomy. Unlike a chatbot that waits for your next prompt, Agentic AI proactively plans, executes, and adapts until the job is done.

Agentic Workflow is an iterative system where an LLM doesn't just talk but acts like planning tasks, web search, code execution, data fetch, and self-correcting errors to achieve a complex goal without human intervention.

AI agents are intelligent systems that are capable of reasoning, planning, deciding and executing multistep tasks to achieve a goal.

An AI Gateway is an intelligent intermediary for managing, securing, and routing interactions between applications and various AI models.

AI hallucination refers to when an AI model confidently generates false, nonsensical, or factually incorrect information or unsupported by evidence.

AI Inference is the process of using a trained AI model to make predictions or decisions on new, unseen data. It is the phase of AI when models stop learning and start producing real-world results

AI tokens are the fundamental units of text that LLMs process to understand and generate text. They are the building blocks that enable AI to interpret prompts and create responses.

Artificial Intelligence (AI) is the capability of machines to perform tasks that typically require human intelligence, like learning and reasoning.

BYOK lets users securely supply their own API keys to AI platforms, with complete control over data, costs, and model choice.

ChatGPT is OpenAI’s conversational interface that allows users to interact with its LLMs to generate text, code, and analysis.

Claude is a family of large language models (LLMs) developed by Anthropic, an AI safety-focused company.

Cursor is an AI-powered code editor that includes advanced AI features like autocomplete, chat, and agent-based automation directly into the coding experience.

DeepSeek AI is a research lab building powerful, open-source LLMs that rival top AI models at significantly lower costs.

Gemini is Google's native multimodal Generative AI assistant for handling text, images, video, audio, and code in conversations via web and apps.

Generative AI is part of Artificial Intelligence that doesn't just analyze existing data, but it creates new data. It can generate text, images, code, audio, and video that has never existed before.

Generative Pre-trained Transformers (GPT) are a family of large language models developed by OpenAI, capable of generating human-like text.

Grok is xAI’s chatbot, designed to deliver direct, often witty responses, with real-time access to public data from the X platform. It is built with fewer content filters and a more candid tone compared to other AI assistants.

Large language models (LLMs) are advanced AI systems trained on massive text data to understand and generate human-like language.

Machine learning is a science of getting computer systems to act without being explicitly programmed. Instead of writing rules, you feed data to an algorithm, and it figures out the rules itself.

Manus AI is an autonomous general-purpose AI agent that executes complex, multi-step tasks on a chat interface similar to ChatGPT and other LLMs.

Mistral AI is a Paris-based startup founded in 2023, specializing in efficient, open-weight LLMs.

MCP is an open standard created by Anthropic that lets AI models connect directly to your data sources and tools without custom integrations.

OpenAI is an AI research and product company known for ChatGPT, which became the first generative AI chat platform to achieve massive global adoption.

Perplexity AI is an answer engine that combines capabilities of large language models with real-time web search to give concise responses with citations.

RAG is a technique that gives an LLM a reference library. It retrieves specific data from your documents and feeds it to the AI so it can answer questions accurately without hallucinating.

Sora is an AI model developed by OpenAI (the creators of ChatGPT) that can create realistic and imaginative videos from text instructions.