Skip to content
AI Tracker
← Back to Agents & METR

The length of tasks AI agents can finish alone has doubled about every 4 months since 2023

Finding

In 2023 the best AI could do, half the time, software tasks taking a skilled person about 4 minutes. By February 2026 it was about 12 hours.

4 min → 12 htasks AI finishes half the time, GPT-4 (2023) → Claude Opus 4.6 (2026)

Length of tasks AI agents can complete on their own

METR time horizon: the length of software tasks (measured in skilled-human time) an AI agent completes with 50% or 80% success. Log scale.

What this measures: How big a job an AI agent can finish on its own. A task's "length" is how long it takes a skilled human; the chart shows the longest tasks an AI completes half the time (blue) or reliably (orange). Tasks are software, machine-learning and cybersecurity work.

  • 50% success
  • 80% success
  • Exponential trend
1 s10 s1 min10 min1 h8 h40 h201920202021202220232024202520262027

Dashed line: exponential fit since 1 Jan 2023, doubling roughly every 4.0 months.

50% success, No longer updated · last data Apr 2026: METR stopped updating this measure on 8 September 2026. The newest models finish almost every task in its test suite, and METR says results above about 16 hours are unreliable, so its last data point is Claude Mythos Preview (April 2026), and no GPT-6, Claude 5.x or Gemini 4 model has a score. METR has moved to measuring how much AI speeds up AI research inside labs. Now watch: METR's AI R&D acceleration audits (e.g. Anthropic, Nov 2026) and Epoch AI/METR's MirrorCode weeks-long coding tasks.

80% success, No longer updated · last data Apr 2026: METR stopped updating this measure on 8 September 2026. The newest models finish almost every task in its test suite, and METR says results above about 16 hours are unreliable, so its last data point is Claude Mythos Preview (April 2026), and no GPT-6, Claude 5.x or Gemini 4 model has a score. METR has moved to measuring how much AI speeds up AI research inside labs. Now watch: METR's AI R&D acceleration audits (e.g. Anthropic, Nov 2026) and Epoch AI/METR's MirrorCode weeks-long coding tasks.

METR Time Horizon 1.1 values (after METR's March 2026 correction), plotted at model release dates. Tasks are well-specified software, ML and cybersecurity tasks. Confidence intervals are wide (e.g. Claude Opus 4.6: 317–3,634 min). METR stopped updating this dataset in September 2026 and says measurements above ~16 hours are unreliable with its current tasks, so models released after spring 2026 are not shown.

Source: METRData as of 9 Nov 2026
Show data table and sources
SeriesDateModelValue
50% success14 Feb 2019GPT-26 s
50% success28 May 2020davinci-002 (GPT-3)6 s
50% success15 Mar 2022GPT-3.5 Turbo Instruct36 s
50% success14 Mar 2023GPT-44 min
50% success6 Nov 2023GPT-4 11064 min
50% success4 Mar 2024Claude 3 Opus4 min
50% success9 Apr 2024GPT-4 Turbo3.7 min
50% success13 May 2024GPT-4o7 min
50% success20 Jun 2024Claude 3.5 Sonnet (Jun)11 min
50% success12 Sept 2024o1-preview20 min
50% success22 Oct 2024Claude 3.5 Sonnet (Oct)21 min
50% success5 Dec 2024o139 min
50% success24 Feb 2025Claude 3.7 Sonnet1 h
50% success16 Apr 2025o32 h
50% success22 May 2025Claude Opus 41.7 h
50% success5 Aug 2025Claude Opus 4.11.7 h
50% success7 Aug 2025GPT-53.4 h
50% success18 Nov 2025Gemini 3 Pro3.7 h
50% success19 Nov 2025GPT-5.1-Codex-Max3.7 h
50% success24 Nov 2025Claude Opus 4.54.9 h
50% success11 Dec 2025GPT-5.25.9 h
50% success5 Feb 2026Claude Opus 4.612 h
50% success5 Feb 2026GPT-5.3-Codex5.8 h
50% success19 Feb 2026Gemini 3.1 Pro6.4 h
50% success5 Mar 2026GPT-5.45.7 h
50% success7 Apr 2026Claude Mythos Preview (limited access)17 h
80% success15 Mar 2022GPT-3.5 Turbo Instruct18 s
80% success14 Mar 2023GPT-454 s
80% success6 Nov 2023GPT-4 110648 s
80% success13 May 2024GPT-4o1.3 min
80% success20 Jun 2024Claude 3.5 Sonnet (Jun)1.7 min
80% success12 Sept 2024o1-preview4.4 min
80% success5 Dec 2024o17.1 min
80% success24 Feb 2025Claude 3.7 Sonnet12 min
80% success16 Apr 2025o330 min
80% success7 Aug 2025GPT-538 min
80% success18 Nov 2025Gemini 3 Pro54 min
80% success11 Dec 2025GPT-5.21.1 h
80% success5 Feb 2026Claude Opus 4.61.2 h
80% success19 Feb 2026Gemini 3.1 Pro1.5 h
80% success7 Apr 2026Claude Mythos Preview (limited access)3.1 h

METR, an independent AI evaluation nonprofit, times how long tasks take skilled people. It then finds the task length at which an AI agent succeeds half the time: its 50% time horizon. Its March 2025 paper found this had doubled roughly every 7 months over six years, and faster recently.

With its updated task set (Time Horizon 1.1, January 2026: 228 tasks, more of them 8+ hours long), METR’s data gives a doubling time of about 129 days, roughly 4 months, for models since 2023 (95% CI 104–158 days). GPT-4 reached about 4 minutes (March 2023), Claude 3.5 Sonnet 20 minutes (October 2024), GPT-5 3.4 hours (August 2025) and Claude Opus 4.6 12 hours (February 2026).

For tasks an agent finishes reliably (80% of the time), the horizon is 4–10× shorter: 70 minutes for Claude Opus 4.6.

What it means

In three years AI agents went from tasks of a few minutes to tasks that take a skilled person a day or more, on tidy, well-defined software problems and with half of attempts failing at that length. For work that must succeed reliably, the figure is closer to an hour. The trend on its own says little about messier real jobs.

Caveats

  • The tasks are well-defined, automatically scored software, machine learning and cybersecurity tasks. METR stresses that real jobs are “messier”.
  • When METR reviewed agent code by hand, none of the pull requests (proposed code changes) were mergeable as-is, even when tests passed.
  • Confidence intervals are wide: for Claude Opus 4.6, 317–3,634 minutes.
  • Human times come from contractors without project context; professionals who know the project would likely be faster.
  • Other domains differ: METR found horizons for visual computer use 40–100× shorter, though growing at similar rates.
Suggest a correction