BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases8

Chess Puzzles leaderboard

As of 6 Nov 2026, the top score on Chess Puzzles is 76.0% by GPT-6 Astra (OpenAI), ahead of GPT-6.1 Sol at 68.0%, across 154 model configurations with a published result (Epoch AI Benchmarking Hub, result dated 3 Sep 2026). Chess puzzles given as a position; the model must find the winning move sequence. Chess 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 Chess Puzzles score at 76.0% (Epoch AI Benchmarking Hub, 3 Sep 2026).
  • GPT-6.1 Sol is second on Chess Puzzles at 68.0% (Epoch AI Benchmarking Hub, 29 Sep 2026).
  • Chess Puzzles: Solving chess puzzles from a board description. Run by Epoch AI (Epoch AI Benchmarking Hub).
  • 154 model configurations have a published Chess Puzzles result on BenchLeader (BenchLeader, 6 Nov 2026).
  • Chess Puzzles does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).

Solving chess puzzles from a board description. Run by Epoch AI.

Published by
Epoch AI Benchmarking Hub
Category
Reasoning
Index weight
Reference only
Models
154
Data as of
6 Nov 2026
Newest result
2 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

Chess puzzles given as a position; the model must find the winning move sequence.

How it is scored

Percent of puzzles solved, run by Epoch AI.

What to keep in mind

A narrow reasoning task with a large model set; good for spotting search-like reasoning.

Top 15
  1. 1GPT-6 Astra (pro)76.0%
  2. 2GPT-6 Astra (max)74.0%
  3. 3GPT-6.1 Sol (max)68.0%
  4. 4GPT-5.6 Sol (pro)68.0%
  5. 5Claude Opus 5.5 (max)67.0%
  6. 6Gemini 3.7 Flash (high)66.0%
  7. 7GPT-5.5 (xhigh)63.0%
  8. 8GPT-5.5 Pro (xhigh)62.0%
  9. 9GPT-5.2 Pro (xhigh)60.0%
  10. 10Gemini 4 Argon (high)59.0%
  11. 11GPT-5.6 Sol (max)59.0%
  12. 12Claude Sonnet 5.5 (max)59.0%
  13. 13GPT-5.4 Pro (xhigh)59.0%
  14. 14Gemini 3.8 Flash (high)58.0%
  15. 15GPT-6 Sol (max)56.0%
154 of 154
#
1GPT-6 Astrapro mode76.0%–2026-09-03
2GPT-6 Astramax74.0%67.82026-09-03
3GPT-6.1 Solmax68.0%67.32026-09-29
4GPT-5.6 Solpro mode68.0%–2026-07-09
5Claude Opus 5.5max67.0%69.32026-09-22
6Gemini 3.7 Flashhigh66.0%62.22026-08-13
7GPT-5.5xhigh63.0%64.32026-04-23
8GPT-5.5 Proxhigh62.0%–2026-04-23
9GPT-5.2 Proxhigh60.0%–2025-12-11
10Gemini 4 Argonhigh59.0%66.02026-09-30
11GPT-5.6 Solmax59.0%65.42026-07-09
12Claude Sonnet 5.5max59.0%65.42026-09-28
13GPT-5.4 Proxhigh59.0%60.02026-03-05
14Gemini 3.8 Flashhigh58.0%62.02026-09-02
15GPT-6 Solmax56.0%63.32026-09-22
16Claude Haiku 5.5max55.0%57.02026-10-07
17Claude Opus 4.8max52.0%60.22026-05-28
18Gemini 3.1 Prohigh51.0%60.12026-02-19
19Claude Fable 5.1max48.0%67.22026-09-01
20GPT-5.4xhigh48.0%63.22026-03-05
21GPT-5.2xhigh48.0%60.02025-12-11
22DeepSeek V4.1 Flashmax48.0%59.02026-09-09
23GPT-5 Prohigh48.0%–2025-10-07
24Claude Fable 5max47.0%65.32026-06-09
25GPT-5.6 Terramax47.0%60.72026-07-09
26DeepSeek V4.1 Flash 1017max46.0%59.22026-10-17
27Qwen3.8 Maxxhigh46.0%57.12026-08-02
28DeepSeek V4 Promax46.0%57.02026-08-13
29Claude Fable 5high45.0%64.32026-06-09
30Gemini 3.5 Flashhigh45.0%62.52026-05-19
31GPT-5.2high45.0%56.02025-12-11
32GPT-5.4 Nanoxhigh45.0%54.02026-03-17
33Claude Sonnet 5max45.0%53.82026-06-30
34Claude Opus 5max43.0%65.42026-07-24
35GPT-5.6 Lunamax43.0%56.32026-07-09
36GPT-5.4high42.0%58.82026-03-05
37GPT-5.1high42.0%57.72025-11-13
38Gemini 3.6 Flashlow42.0%–2026-07-21
39Gemini 3.5 Flashlow42.0%–2026-05-19
40Claude Opus 4.7max41.0%59.52026-04-16
41Grok 4.7xhigh41.0%58.32026-09-21
42Gemini 3.6 Flashhigh40.0%61.22026-07-21
43Gemini 3 Flashhigh40.0%58.32025-12-17
44NewDoubao Seed Evolvinghigh40.0%–2026-11-02
45Muse Spark 1.3max37.0%61.92026-09-02
46GLM 5.3max37.0%59.02026-08-14
47GPT-5.4medium37.0%54.72026-03-05
48o3medium37.0%53.32025-04-16
49Kimi K3high37.0%–2026-07-16
50Kimi K3max36.0%62.12026-07-16
51GPT-5high36.0%57.92025-08-07
52Claude Opus 4.7xhigh36.0%57.42026-04-16
53GPT-6 Lunamax36.0%56.72026-09-22
54GPT-5.2medium36.0%56.22025-12-11
55o3high36.0%53.82025-04-16
56GPT-5.4 Nanohigh36.0%48.32026-03-17
57o3low36.0%46.32025-04-16
58Grok 4.5high35.0%56.92026-07-08
59Muse Spark 1.1xhigh34.0%59.52026-07-09
60Muse Spark 1.3xhigh34.0%58.52026-09-02

More reasoning benchmarks · All LLM benchmarks

Related benchmarks

Top head-to-heads on Chess Puzzles

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

Chess Puzzles: questions

What does Chess Puzzles measure?
Chess puzzles given as a position; the model must find the winning move sequence. Scores are reported in percent of tasks solved; higher is better.
What is the top score on Chess Puzzles?
As of 6 Nov 2026, the highest score on Chess Puzzles is 76.0% by GPT-6 Astra (result dated 3 Sep 2026), ahead of GPT-6.1 Sol at 68.0%. Scores come from Epoch AI Benchmarking Hub and are re-read daily.
Which AI model leads Chess Puzzles?
GPT-6 Astra leads Chess Puzzles with 76.0% as of 6 Nov 2026 (result dated 3 Sep 2026), ahead of GPT-6.1 Sol at 68.0%.
How many models have Chess Puzzles results?
154 model configurations have a Chess Puzzles result on BenchLeader, all taken from Epoch AI Benchmarking Hub.
Who runs Chess Puzzles and how often is it updated?
Chess 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 Chess Puzzles count toward the BenchLeader Index?
No. Chess Puzzles is shown for reference but left out of the composite index.