BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases7

Mystery Game Puzzles leaderboard

As of 6 Nov 2026, the top score on Mystery Game Puzzles is 84.0% by GPT-6 Astra (OpenAI), ahead of GPT-6.1 Sol at 80.0%, across 143 model configurations with a published result (Epoch AI Benchmarking Hub, result dated 3 Sep 2026). Deduction puzzles with clues and suspects; the model must reason to the unique solution. Mystery Game Puzzles is shown for reference only and does not count toward the BenchLeader Index.

Key facts (as of 6 Nov 2026)

  • GPT-6 Astra holds the top Mystery Game Puzzles score at 84.0% (Epoch AI Benchmarking Hub, 3 Sep 2026).
  • GPT-6.1 Sol is second on Mystery Game Puzzles at 80.0% (Epoch AI Benchmarking Hub, 29 Sep 2026).
  • Mystery Game Puzzles: Deduction puzzles in the style of mystery games. Run by Epoch AI (Epoch AI Benchmarking Hub).
  • 143 model configurations have a published Mystery Game Puzzles result on BenchLeader (BenchLeader, 6 Nov 2026).
  • Mystery Game Puzzles does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).

Deduction puzzles in the style of mystery games. Run by Epoch AI.

Published by
Epoch AI Benchmarking Hub
Category
Reasoning
Index weight
Reference only
Models
143
Data as of
6 Nov 2026
Newest result
29 Oct 2026

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

What the test looks like

Deduction puzzles with clues and suspects; the model must reason to the unique solution.

How it is scored

Percent solved, run by Epoch AI.

What to keep in mind

Puzzle styles are specific; models can be tuned to them.

Top 15
  1. 1GPT-6 Astra (max)84.0%
  2. 2GPT-6.1 Sol (max)80.0%
  3. 3GPT-6.1 Terra (max)74.0%
  4. 4Claude Opus 5.5 (max)71.0%
  5. 5Claude Sonnet 5.5 (max)65.0%
  6. 6Gemini 4 Argon (high)64.0%
  7. 7Claude Opus 5 (max)59.0%
  8. 8Claude Fable 5.1 (max)58.0%
  9. 9GPT-5.6 Sol (max)58.0%
  10. 10GPT-5.5 (xhigh)56.0%
  11. 11GPT-6 Sol (max)56.0%
  12. 12Claude Fable 5 (max)52.0%
  13. 13GPT-5.5 (high)52.0%
  14. 14Muse Wonder (max)51.0%
  15. 15DeepSeek V4.1 Flash 1017 (max)49.0%
143 of 143
#
1GPT-6 Astramax84.0%67.82026-09-03
2GPT-6.1 Solmax80.0%67.32026-09-29
3GPT-6.1 Terramax74.0%65.32026-10-26
4Claude Opus 5.5max71.0%69.32026-09-22
5Claude Sonnet 5.5max65.0%65.42026-09-28
6Gemini 4 Argonhigh64.0%66.02026-09-30
7Claude Opus 5max59.0%65.42026-07-24
8Claude Fable 5.1max58.0%67.22026-09-01
9GPT-5.6 Solmax58.0%65.42026-07-09
10GPT-5.5xhigh56.0%64.32026-04-23
11GPT-6 Solmax56.0%63.32026-09-22
12Claude Fable 5max52.0%65.32026-06-09
13GPT-5.5high52.0%63.72026-04-23
14Muse Wondermax51.0%62.32026-10-12
15DeepSeek V4.1 Flash 1017max49.0%59.22026-10-17
16Gemini 3.8 Flashhigh47.0%62.02026-09-02
17DeepSeek V4.1 Flashmax43.0%59.02026-09-09
18DeepSeek V4 Promax43.0%57.02026-08-13
19Qwen3.8 Maxxhigh38.0%57.12026-08-02
20Claude Opus 537.0%65.22026-07-24
21GPT-5.4xhigh37.0%63.22026-03-05
22Gemini 3.7 Flashhigh37.0%62.22026-08-13
23Claude Opus 4.8max36.0%60.22026-05-28
24GPT-5.6 Terramax35.0%60.72026-07-09
25Claude Sonnet 5max35.0%53.82026-06-30
26Gemini 3.1 Prohigh34.0%60.12026-02-19
27Grok 4.6xhigh34.0%57.92026-08-12
28DeepSeek V4 Flashmax34.0%57.02026-07-31
29GLM 5.3max33.0%59.02026-08-14
30GPT-5.6 Solno reasoning33.0%52.22026-07-09
31Gemini 3.5 Flashhigh32.0%62.52026-05-19
32Qwen3.7max32.0%60.32026-05-19
33Gemini 3.1 Promedium32.0%–2026-02-19
34Claude Opus 4.8xhigh31.0%–2026-05-28
35Gemini 3.6 Flashhigh30.0%61.22026-07-21
36Claude Haiku 5.5max30.0%57.02026-10-07
37Grok 4.7xhigh29.0%58.32026-09-21
38o3high29.0%53.82025-04-16
39Gemini 3.1 Prolow29.0%–2026-02-19
40Claude Opus 4.7max28.0%59.52026-04-16
41GPT-5.5low28.0%58.22026-04-23
42GPT-5.4medium28.0%54.72026-03-05
43Gemini 3.5 Flashlow28.0%–2026-05-19
44Kimi K3max26.0%62.12026-07-16
45GPT-5.6 Sollow26.0%60.42026-07-09
46Gemini 3 Flashlow26.0%–2025-12-17
47Qwen3.7no reasoning26.0%–2026-05-19
48Muse Spark 1.3max25.0%62.02026-09-02
49Claude Opus 4.6max25.0%55.82026-02-05
50Gemini 3 Flashminimal25.0%53.92025-12-17
51Gemini 3.6 Flashminimal25.0%–2026-07-21
52GPT-5high23.0%57.92025-08-07
53GPT-5.2high23.0%56.02025-12-11
54o3medium23.0%53.32025-04-16
55Gemini 3.6 Flashlow23.0%–2026-07-21
56Claude Opus 4.522.0%57.92025-11-24
57GPT-5.2medium22.0%56.22025-12-11
58Qwen3.6 35B A3Bno reasoning22.0%44.42026-04-14
59GPT-5.6 Lunamax21.0%56.32026-07-09
60Claude Opus 4.121.0%51.42025-08-05

More reasoning benchmarks · All LLM benchmarks

Related benchmarks

Top head-to-heads on Mystery Game Puzzles

Cite as: BenchLeader, “Mystery Game Puzzles leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/mystery_game_puzzles.

Mystery Game Puzzles: questions

What does Mystery Game Puzzles measure?
Deduction puzzles with clues and suspects; the model must reason to the unique solution. Scores are reported in percent of tasks solved; higher is better.
What is the top score on Mystery Game Puzzles?
As of 6 Nov 2026, the highest score on Mystery Game Puzzles is 84.0% by GPT-6 Astra (result dated 3 Sep 2026), ahead of GPT-6.1 Sol at 80.0%. Scores come from Epoch AI Benchmarking Hub and are re-read daily.
Which AI model leads Mystery Game Puzzles?
GPT-6 Astra leads Mystery Game Puzzles with 84.0% as of 6 Nov 2026 (result dated 3 Sep 2026), ahead of GPT-6.1 Sol at 80.0%.
How many models have Mystery Game Puzzles results?
143 model configurations have a Mystery Game Puzzles result on BenchLeader, all taken from Epoch AI Benchmarking Hub.
Who runs Mystery Game Puzzles and how often is it updated?
Mystery Game Puzzles is published by Epoch AI Benchmarking Hub. BenchLeader re-reads the published results once a day and records the date each result was published.
Does Mystery Game Puzzles count toward the BenchLeader Index?
No. Mystery Game Puzzles is shown for reference but left out of the composite index.