long-horizon

Long-horizon is an AI term for tasks or reasoning that span many steps over an extended period, tied to an agent's ability to plan and follow through.

7 articlesLast mentioned

Long-horizon is a term used in AI for tasks that must be carried out over many steps and an extended period (long-horizon tasks) and for the reasoning such tasks require (long-horizon reasoning). Unlike a single question and answer, these tasks require breaking a goal into sub-steps, keeping track of intermediate results and correcting errors along the way until completion. The expression originally referred to reinforcement learning problems in which rewards arrive far in the future.

As AI agents take on lengthy work such as software development, computer operation and research, long-horizon capability has become a key measure of model performance. Because small errors accumulate as the number of steps grows, the length of task a model can complete autonomously is sometimes used as an evaluation metric.

This entry is based on AIPOST articles and widely known facts. If something is wrong, please send us a correction request.

Articles covering this entry

  • Anthropic's Always-On Agent Guide: Five Primitives and a First Agent to BuildAI in Practice ·

    Anthropic reports that on a long-horizon coding benchmark, Claude Opus 5.5 costs 78% less per solved task than Claude Opus 5, compared with an average reduction…

    Read article
  • Claude Fable 5.2 Benchmarks and Five Real Tasks: Where the Gains Show UpAI Performance ·

    …ad over Opus 5.5 and GPT-6 Astra is consistent across coding, science and long-horizon knowledge work, and the hands-on tasks back that up for strategy-heavy jo…

    Read article
  • METR Report: AI Progress Doubles in Speed, Raising 2027 Self-Improvement DebateAI News ·

    …METR has used: human software engineers and AI systems tested on complex, long-horizon tasks, with the question being whether the AI completes economically valu…

    Read article
  • Anthropic IPO Valuation: Why Cost of Capital Isn't the Weak LinkAI News ·

    A valuation of Anthropic will be wrong in hindsight; every long-horizon forecast is.

    Read article
  • Compound's Multiplayer Email Agents and Headless Office Engine for Deal TeamsAI Startups ·

    Existing commercial tools, in Compound's assessment, still struggle with long-horizon generative work like producing a 50-page committee deck from scratch.

    Read article
  • GPT-6.1 Terra, Dot Texting and Gemini 4 Carbon: What Works Now and What WaitsAI News ·

    …it as its strongest model so far with state-of-the-art cybersecurity and long-horizon reasoning, the ability to carry a task through many steps over a long str…

    Read article
  • OpenAI's ChatGPT Lead Says to Build for Models 10x Better Within a YearAI News ·

    Dots combines more than two years of research on long-horizon tasks and persistent memory with the Codex execution harness and continuous background operation.

    Read article

Found a mistake or missing an entry?

Wiki entries are based on AIPOST articles and widely known facts. If an entry is wrong, send us a correction request and we will check and fix it. Names from new articles are added to the wiki after the article is published.