ReAct Pattern

Full definition

Introduced by Yao et al. (Princeton/Google, 2022), ReAct outperformed act-only prompting on every benchmark tested and was competitive with chain-of-thought on knowledge-intensive tasks: ahead on Fever, slightly behind on HotpotQA, with the two combined performing best. On the ALFWorld and WebShop decision-making benchmarks it beat imitation- and reinforcement-learning methods by 34 and 10 percentage points. The pattern is now the canonical agent loop in LangChain, LlamaIndex, and most agent frameworks. Modern function-calling LLMs implement ReAct natively in their tool-use training.

Why it matters

ReAct is the simplest pattern that turns a chat LLM into an agent. Understanding it is prerequisite to designing reliable agentic workflows — including knowing when ReAct's chattiness is excessive overhead and a planner-executor split would be cheaper.

Example

Thought: "I need yesterday's closing price." Action: get_stock_price("AAPL", "2026-05-01"). Observation: 224.18. Thought: "Now I can compute the daily change."

Next steps

• Get a free workflow map →We map one of your workflows and give you a written verdict: software, a process change or AI. No obligation.
• AI consulting cost guide →Hourly rates and project budgets by firm tier, from independents to the Big Four.

Related terms

• Chain-of-Thought (CoT)
  Chain-of-Thought (CoT) is a prompting and training technique in which a language model is encouraged to reason step-by-step before producing its final answer — improving accuracy on math, logic, and multi-hop reasoning in large models (in Wei et al., 2022, GSM8K math accuracy more than doubled for the largest GPT and PaLM models).
• Agentic Workflow
  An agentic workflow is a multi-step AI process where a model plans, takes actions through tools, observes results, and iterates toward a goal — replacing single-shot prompting with a loop that reasons, acts, and self-corrects.
• Tool Use (LLM)
  LLM tool use is the pattern in which a language model acts through external functions, APIs, databases or code execution: the model requests a tool, the host application runs it and returns the result, and the model continues, turning a text generator into a controller that can act. Function calling is the structured-output mechanism most tool use is built on.
• AI Agent / Agentic AI
  An AI agent is an LLM-driven system that, given a goal, plans a sequence of steps, calls tools or APIs, observes the results, and iterates until the goal is reached — going beyond single-turn chat to multi-step autonomous action.

More training & optimization terms

• Self-Consistency
  Self-consistency is a prompting technique that samples multiple chain-of-thought reasoning paths from an LLM at non-zero temperature, then selects the most common final answer by majority vote — improving reasoning accuracy at the cost of more tokens.
• Tree of Thoughts
  Tree of Thoughts (ToT) is a reasoning framework where an LLM explores multiple reasoning branches as a tree, generating intermediate "thoughts," evaluating partial states with the model itself, and backtracking — turning generation into deliberate search.
• LoRA (Low-Rank Adaptation)
  LoRA (Low-Rank Adaptation) is a parameter-efficient fine-tuning method that freezes a pretrained model's weights and injects small trainable low-rank matrices into selected layers — cutting trainable parameters by up to 10,000x and GPU memory by about 3x with negligible quality loss on most tasks.