Skip to content
Last Dictionary
Plane
Agents & HarnessesLanguage Models & TransformersTraining & Post-training

tool use

nounAlso: tool learning, tool-augmented LLM, tool-using agent

In plain wordsAn AI model's ability to reach beyond its own knowledge by operating outside tools — searching the web, running code, using a calculator or an app — and folding the results into its work.

  1. The capability of a large language modellarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code. to accomplish tasks by deciding when to invoke external toolstool, n.: Any outside capability an AI model is allowed to use — a web search, a calculator, a code runner, a database or an app — described to the model so it can ask for it by name., choosing which tool and arguments, interpreting the results and continuing — the foundation of AI agentsAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done.. Tool use compensates for well-known weaknesses of models (stale parametric knowledgeparametric knowledge, n.: What an AI model 'knows' from training, stored in the numbers of its network rather than looked up in documents — like facts you have memorized rather than ones you check in a book., unreliable arithmetic, no access to private data, no ability to act) and turns text generation into action. It is taught by supervised fine-tuningsupervised fine-tuning, n.: Teaching a pretrained AI model by showing it many examples of good answers to requests, so it learns to imitate them — like learning a job by studying worked examples. on demonstrations, by self-supervised schemes such as ToolformerToolformer, prop. n.: A 2023 Meta research model that learned on its own when a tool — a calendar, a translator, a search engine, a calculator — would help it predict text, keeping only the tool uses that actually paid off., and increasingly by reinforcement learningreinforcement learning, n.: A way of teaching a computer program by trial and error: it tries actions, earns points for good results and penalties for bad ones, and gradually learns which choices pay off over time. in environments where tool calls affect outcomes; at inference time it is exposed through function callingfunction calling, n.: A feature of AI model services that lets the model reply with a neatly formatted request to run one of a program's functions, rather than with ordinary text; the program, not the model, then runs it. interfaces and protocols such as the Model Context ProtocolModel Context Protocol, prop. n.: An open standard that lets AI assistants plug into outside apps and data — files, calendars, databases, business software — through one common connector instead of a custom link for every pairing; often likened to a USB-C port for AI.. Failure modes include unnecessary calls, invented tools or arguments, misread outputs, and obeying malicious instructions embedded in tool results (indirect prompt injectionindirect prompt injection, n.: A prompt-injection attack in which the trick instructions are hidden in content the AI reads on its own — a web page, an email, a file — rather than typed by the person using it.).

    With tool use enabled, the model checks the exchange rate instead of guessing it.

Origin

Tool use has long been studied as a mark of animal and human intelligence. For language models, systems such as WebGPTWebGPT, prop. n.: A 2021 OpenAI research system that answered questions by searching and browsing the web in a simple text browser, collecting sources and citing them — an early ancestor of today's research agents. (2021), MRKLMRKL system, n.: An early design, pronounced 'miracle', in which a language model acts as a switchboard, sending each part of a question to the right specialist — a calculator, a database, a search tool — and combining the answers. and TALM (2022) called external tools; ToolformerToolformer, prop. n.: A 2023 Meta research model that learned on its own when a tool — a calendar, a translator, a search engine, a calculator — would help it predict text, keeping only the tool uses that actually paid off. (Timo Schick et al., February 2023) showed a model could learn when to call APIs, and Andrew Ng listed tool use among four agentic design patterns in March 2024.

Related

See also
  • tooltool, n.: Any outside capability an AI model is allowed to use — a web search, a calculator, a code runner, a database or an app — described to the model so it can ask for it by name.
  • tool calltool call, n.: A single request from an AI model to use one of its tools — naming the tool and filling in the details — which the surrounding software then carries out and reports back on.
  • function callingfunction calling, n.: A feature of AI model services that lets the model reply with a neatly formatted request to run one of a program's functions, rather than with ordinary text; the program, not the model, then runs it.
  • ToolformerToolformer, prop. n.: A 2023 Meta research model that learned on its own when a tool — a calendar, a translator, a search engine, a calculator — would help it predict text, keeping only the tool uses that actually paid off.
  • AI agentAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done.
  • code executioncode execution, n.: Letting an AI write a small program and actually run it — to do math, analyze a spreadsheet or test its own code — then use the real output instead of guessing.
  • indirect prompt injectionindirect prompt injection, n.: A prompt-injection attack in which the trick instructions are hidden in content the AI reads on its own — a web page, an email, a file — rather than typed by the person using it.

Sources

  1. Schick, T. et al. (2023). Toolformer: Language Models Can Teach Themselves to Use Tools. arXiv:2302.04761. arxiv.org/abs/2302.04761v1
  2. Mialon, G. et al. (2023). Augmented Language Models: a Survey. arXiv:2302.07842. arxiv.org/abs/2302.07842

Mentioned in

AI agentAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done., agentic RAGagentic RAG, n.: Retrieval in which the AI runs its own research: it decides what to search for, looks at the results, searches again if needed, and only then answers., agentic workflowagentic workflow, n.: A multi-step process in which AI models draft, check, use tools and revise their work over several rounds, instead of answering in one shot; often the steps are laid out by developers in advance., AI assistantAI assistant, n.: An AI program, such as ChatGPT, Claude or Gemini, that helps a person through conversation: answering questions, writing, summarizing, coding and, increasingly, taking actions on their behalf., AI engineeringAI engineering, n.: The craft of building useful, reliable products on top of AI models — choosing models, writing prompts, connecting data and tools, testing quality and running it all safely at scale., BrowseCompBrowseComp, prop. n.: A test of AI web-research agents: 1,266 puzzle-like questions whose answers take a long, persistent online hunt to uncover, though each answer is short and simple to confirm., code executioncode execution, n.: Letting an AI write a small program and actually run it — to do math, analyze a spreadsheet or test its own code — then use the real output instead of guessing., conversational AIconversational AI, n.: The technology that lets computers hold conversations with people in everyday language, by text or voice. It powers chatbots, voice assistants and automated customer-service lines., fancy autocompletefancy autocomplete, n.: A dismissive description of AI chatbots as nothing more than a souped-up version of the feature on a phone keyboard that suggests your next word., function callingfunction calling, n.: A feature of AI model services that lets the model reply with a neatly formatted request to run one of a program's functions, rather than with ordinary text; the program, not the model, then runs it., GAIAGAIA, prop. n.: A test of AI assistants made of real-world questions that people can answer with a browser and some patience, but that require an AI to search, read files and combine facts., gpt-ossgpt-oss, prop. n.: Two free-to-download reasoning models released by OpenAI in August 2025 — its first downloadable language models since 2019 — that anyone may run, modify and use commercially on their own hardware., harnessharness, n.: The software around an AI model that gives it instructions and tools, carries out its actions and decides when it stops. The model is the engine; the harness is the rest of the car., inference providerinference provider, n.: A company that runs AI models on its own computers and sells access to them over the internet, charging by use, so customers need not buy or operate the hardware themselves., Moonshot AIMoonshot AI, prop. n.: A Beijing artificial intelligence company, founded in 2023, that makes the Kimi chatbot and very large models that anyone can download., neuro-symbolic AIneuro-symbolic AI, n.: AI that combines neural networks, which learn patterns from data, with symbolic methods, which follow explicit rules and logic, aiming for systems that both learn flexibly and reason reliably., ReActReAct, prop. n.: A method in which an AI alternates between thinking out loud about what to do next and taking an action, such as a web search, then uses what it observed to plan its next step., scaffoldingscaffolding, n.: The code, prompts and tools built around an AI model so it can carry out tasks it could not manage from a bare prompt — like the scaffolding that lets builders reach every part of a building., symbol grounding problemsymbol grounding problem, n.: The puzzle of how words or symbols inside a machine can mean anything by themselves, rather than only pointing to more symbols — like a dictionary that defines every word using other words., τ-benchτ-bench, prop. n.: A test of AI customer-service agents in which a simulated customer, played by another AI, chats with the agent, which must use company tools and follow written rules to resolve the request., tooltool, n.: Any outside capability an AI model is allowed to use — a web search, a calculator, a code runner, a database or an app — described to the model so it can ask for it by name., voice agentvoice agent, n.: An AI system you talk to out loud, for example on a phone call, that listens, answers in a natural-sounding voice and can carry out tasks such as booking appointments or checking an order., AgentDojoAgentDojo, prop. n.: A testing environment that measures how easily AI assistants that use email, calendars or banking tools can be hijacked by hidden instructions planted in the content they read., agentic reinforcement learningagentic reinforcement learning, n.: Training an AI assistant by rewarding success on long, multi-step jobs where it must use tools, browse, run code and react to what happens, rather than answer in one go.