BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases7

METR Time Horizons leaderboard

As of 6 Nov 2026, the top score on METR Time Horizons is 85.2% by Claude Mythos Preview (Anthropic), ahead of Claude Opus 4.6 at 78.9%, across 40 model configurations with a published result (METR, result dated 7 Apr 2026). Software and research tasks of increasing length; METR fits the task duration at which a model succeeds half the time. METR Time Horizons is shown for reference only and does not count toward the BenchLeader Index.

Key facts (as of 6 Nov 2026)

  • Claude Mythos Preview holds the top METR Time Horizons score at 85.2% (METR, 7 Apr 2026).
  • Claude Opus 4.6 is second on METR Time Horizons at 78.9% (METR, 5 Feb 2026).
  • METR Time Horizons: METR's task suite behind the 50%-time-horizon measure (METR).
  • 40 model configurations have a published METR Time Horizons result on BenchLeader (BenchLeader, 6 Nov 2026).
  • METR Time Horizons does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).

METR's task suite behind the 50%-time-horizon measure.

Published by
METRdata via Epoch AI Benchmarking Hub
Category
Agents & tools
Index weight
Reference only
Models
40
Data as of
6 Nov 2026

CC BY 4.0 — Epoch AI, ‘AI Benchmarking Hub’, epoch.ai/benchmarks. Mirrored boards credit their original publishers.

What the test looks like

Software and research tasks of increasing length; METR fits the task duration at which a model succeeds half the time.

How it is scored

Average success across the task suite, as published by METR.

What to keep in mind

The headline METR number is a fitted time horizon; this is the underlying success rate.

Top 15
  1. 1Claude Mythos Preview85.2%
  2. 2Claude Opus 4.678.9%
  3. 3Gemini 3.1 Pro77.0%
  4. 4GPT-5.2 (high)75.3%
  5. 5Claude Opus 4.575.0%
  6. 6GPT-5.3-Codex74.5%
  7. 7GPT-5.4 (xhigh)74.3%
  8. 8Gemini 3 Pro71.0%
  9. 9GPT-5.1-Codex-Max (max)70.8%
  10. 10GPT-5 (medium)69.6%
  11. 11GPT-5 (high)69.4%
  12. 12Claude Sonnet 4.567.4%
  13. 13Claude Opus 4.166.8%
  14. 14Grok 466.6%
  15. 15o3 (medium)65.4%
40 of 40
#
1Claude Mythos Preview85.2%–2026-04-07
2Claude Opus 4.678.9%62.32026-02-05
3Gemini 3.1 Pro77.0%60.52026-02-19
4GPT-5.2high75.3%56.02025-12-11
5Claude Opus 4.575.0%57.92025-11-24
6GPT-5.3-Codex74.5%–2026-02-05
7GPT-5.4xhigh74.3%63.22026-03-05
8Gemini 3 Pro71.0%60.22025-11-18
9GPT-5.1-Codex-Maxmax70.8%–2025-11-19
10GPT-5medium69.6%58.12025-08-07
11GPT-5high69.4%57.92025-08-07
12Claude Sonnet 4.567.4%53.72025-09-29
13Claude Opus 4.166.8%51.42025-08-05
14Grok 466.6%58.32025-07-09
15o3medium65.4%53.32025-04-16
16Claude Opus 463.9%52.62025-05-22
17o4-minimedium63.9%51.82025-04-16
18o363.6%61.02025-04-16
19Claude Sonnet 462.0%50.52025-05-22
20Claude 3.7 Sonnet60.0%49.92025-02-24
21Kimi K2 Thinkingthinking59.2%53.22025-11-06
22gpt-oss-120b56.6%45.52025-08-05
23Gemini 2.5 Pro55.4%54.52025-06-05
24DeepSeek R1 052853.8%49.32025-05-28
25DeepSeek R151.9%47.32025-01-20
26o1medium51.0%48.12024-12-17
27DeepSeek V347.4%44.42024-12-26
28Claude 3.5 Sonnet45.2%47.12024-10-22
29o145.1%53.42024-09-12
30GPT-4o40.8%43.32024-05-13
31GPT-4 Turbo36.7%41.22024-04-09
32GPT-436.1%42.42023-03-14
33Qwen2.5 72B35.8%43.22024-09-19
34GPT 4 012535.2%–2024-01-25
35Qwen2-72B29.9%41.32024-06-07
36Claude 3 Opus29.5%39.02024-02-29
37GPT-4 110628.9%–2023-11-06
38GPT-3.5-turbo21.5%–2023-09-18
39Davinci 00216.2%––
40Gpt2 Xl10.1%–2019-11-05

More agents & tools benchmarks · All LLM benchmarks

Related benchmarks

Top head-to-heads on METR Time Horizons

Cite as: BenchLeader, “METR Time Horizons leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/metr_time_horizons.

METR Time Horizons: questions

What does METR Time Horizons measure?
Software and research tasks of increasing length; METR fits the task duration at which a model succeeds half the time. Scores are reported in percent of tasks solved; higher is better.
What is the top score on METR Time Horizons?
As of 6 Nov 2026, the highest score on METR Time Horizons is 85.2% by Claude Mythos Preview (result dated 7 Apr 2026), ahead of Claude Opus 4.6 at 78.9%. Scores come from METR and are re-read daily.
Which AI model leads METR Time Horizons?
Claude Mythos Preview leads METR Time Horizons with 85.2% as of 6 Nov 2026 (result dated 7 Apr 2026), ahead of Claude Opus 4.6 at 78.9%.
How many models have METR Time Horizons results?
40 model configurations have a METR Time Horizons result on BenchLeader, all taken from METR via Epoch AI Benchmarking Hub.
Who runs METR Time Horizons and how often is it updated?
METR Time Horizons is published by METR. BenchLeader re-reads the published results once a day and records the date each result was published.
Does METR Time Horizons count toward the BenchLeader Index?
No. METR Time Horizons is shown for reference but left out of the composite index.