METR 50% time horizon

faster than normal80/95leading

1What this measures

Length (in human-expert minutes) of software tasks a frontier model completes with 50% success, per METR's Horizon v1.1 suite, plotted by model release date. The band applies to the doubling time of that series over the most recent window METR itself reports (2024 onward), recomputed here from METR's corrected data file; METR's published 2023-onward figure is shown alongside.

Why it matters. The cleanest single series for the methods stock. If capability doubles every few months the invention valve is wide open; the paper's claim is that this alone does not move the later stocks.

Proxy types: benchmark
Unit: minutes
Cadence: per release

2How we track this

series metr.*.horizon_50.pt
series metr.suite.doubling_time_days.pt
series metr_blog.suite.doubling_time_days_since_2024.pt
series metr_blog.suite.doubling_time_days_since_2023.pt
series metr_blog.suite.doubling_time_days_2019_2025.pt
source METR time horizons (v1.1) · default tier 1 · METR; analysis code MIT at https://github.com/METR/eval-analysis-public
source METR research posts · default tier 6 · METR

Normal band: ≥ 213 days
Fast band: ≤ 122 days
Falsifying: —

Normal = doubling time ≥ 7 months (≈213 days, METR's 2019–2025 trend); fast = ≤ 4 months (≈122 days), the AI 2027 assumption. Between the two is `emerging`. The bands are about the current pace, so the input is a log-linear fit over the 2024-onward window, the most recent window METR reports (89 days in the TH1.1 post of 29 Jan 2026), recomputed from METR's corrected data. On the 2023-onward window METR's published figure is 128.7 days, in the gap between the bands: the reading is window-sensitive and both numbers are shown.

Applied to metric:horizon_doubling_days_50_since_2024.

3Tracker interpretation

Capability is moving fast by any historical standard. Whether it is faster than the AI-2027 doubling assumption is the live question the band answers.

4Evidence

Latest point

17.4 h(8.5 h–55.1 h)⚑as of 2026-04-07

claude_mythos_preview_early_inspect

Value the bands apply to

108 days(99 days–119 days)as of 2026-03-05(25 obs)

30 observations. Hollow points are disputed (see counterevidence). Every point links to its observation.

Model comparison (lower AIC fits better; both on ln(horizon) residuals)

fit	estimate	95% interval	n	R²	AIC	inputs
`horizon_doubling_days_50_since_2024`	108 days	99 days–119 days	20	0.966	-44.8	obs:0d21fe65 obs:19631991+23 more
`horizon_doubling_days_50_since_2023`	125 days	111 days–143 days	22	0.933	-31.0	obs:0d21fe65 obs:19631991+23 more
`horizon_blowup_year_50_since_2024`	2025.8	2025.4–2026.8	20	0.645	78.1	obs:0d21fe65 obs:19631991+23 more

5Status and reasoning

faster than normalsince 2026-09-10 · claude-initial-seed

Band input switched to the most recent window METR reports (2024 onward), per the updated rationale: our log-linear fit on METR's corrected v1.1 data gives a doubling time of 108 days, inside the fast band (<=122). METR's own TH1.1 post (29 Jan 2026) reports 89 days since 2024. On the 2023-onward window the figures are 128.7 (METR) and 125 (ours), which would read emerging; the status is window-sensitive and both are shown on the page.

6Timeline notes

2026-04-07 claude_mythos_preview_early_inspect · 17.4 h(8.5 h–55.1 h)⚑disputedas of 2026-04-07
2026-03-05 gpt_5_4 · 5.7 h(3.1 h–12.8 h)as of 2026-03-05
2026-02-19 gemini_3_1_pro · 6.4 h(3.9 h–11.6 h)as of 2026-02-19
2026-02-05 gpt_5_3_codex · 5.8 h(3.2 h–13.6 h)as of 2026-02-05
2026-02-05 claude_opus_4_6_inspect · 12.0 h(5.3 h–60.6 h)as of 2026-02-05
2025-12-11 gpt_5_2 · 5.9 h(3.3 h–13.6 h)as of 2025-12-11

7Counterevidence

What cuts against this reading

On the 2023-onward window (METR's published 128.7 days, our fit 125) the status would be `emerging`, not `faster_than_normal`; the status depends on the window. METR notes measurements above 16 hours are unreliable with the current task suite, so the newest points carry wide intervals and are flagged disputed, and human baselines exist for only 5 of the 31 tasks over 8 hours. Horizon is a software-task construct and may not transfer to other work.

8Update history

2026-09-10emerging to faster than normalconf 80 → 80 · claude-initial-seed
Band input switched to the most recent window METR reports (2024 onward), per the updated rationale: our log-linear fit on METR's corrected v1.1 data gives a doubling time of 108 days, inside the fast band (<=122). METR's own TH1.1 post (29 Jan 2026) reports 89 days since 2024. On the 2023-onward window the figures are 128.7 (METR) and 125 (ours), which would read emerging; the status is window-sensitive and both are shown on the page.
obs:0d21fe65 obs:19631991 obs:261c1f51 obs:2eb7c897 obs:3466aea4 obs:37e97a6f+14 more
2026-09-10unmeasured to emergingconf — → 80 · claude-initial-seed
METR's published 2023-onward doubling time is 128.7 days (CI 104-158), between the normal band (>=213 days) and the fast band (<=122 days). Initial seed; the brief expected faster_than_normal. Pending Alex's review.
obs:479bfe0e

9Confidence

80 / 95 — good evidence, some ambiguity

Confidence is independent of status: 90–95 multiple strong independent sources; 70–89 good evidence, some ambiguity; 50–69 mixed or hard to operationalise; below 50 limited or vague.

10Related

METR 80% time horizon faster than normal
50%/80% horizon ratio consistent with normal
Bottlenecks #1, #2, #3 (Narayanan & Kapoor's list)
Crosswalk: Methods ⇄ Model (same valve)