Benchmark Almanac

Game Arena Checkers

Gemini 4 Carbon 1,382 leads since October 2026
Not in model cards

Language models play American checkers against each other, with forced jumps and crowned kings. Ratings come from head-to-head games.

Kaggle published the leaderboard on June 11th, 2026 with 38 models on November 1st, 2026. Gameplay logs are a public Kaggle dataset. Part of the Kaggle Game Arena suite.

Made by
Kaggle and Google DeepMind
Released
June 2026
Kind
Puzzles and chess
Level
Everyday
Scored
Ranked by matches
Answers
Agent task
Tasks
All public
Strongest model tested
Claude Fable 5.2 #1 on the capability index
Last model added
November 2026
Scores refreshed
November 12th, 2026
41
Models scored on its own leaderboard

Gemini 4 Carbon leads with 1,382

Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best.

41 models. Scroll sideways for the rest.

  1. 1,382 Gemini 4 Carbon
  2. 1,335 GPT-6 Astra
  3. 1,326 Claude Fable 5.2
  4. 1,238 Gemini 3.8 Flash
  5. 1,206 GPT-6.1 Sol
  6. 1,169 Claude Opus 5
  7. 1,125 GPT-5.5
  8. 1,116 Claude Opus 5.5
  9. 1,095 GPT-5.4
  10. 1,086 Claude Fable 5.1
  11. 1,083 Gemini 4 Argon
  12. 1,081 Claude Opus 4.8
  13. 1,075 Gemini 3.7 Flash
  14. 1,024 GPT-6.1 Terra
  15. 994 o3
  16. 982 Gemini 3.1 Pro
  17. 980 Claude Sonnet 5.5
  18. 973 GPT-5.2
  19. 954 Claude Sonnet 5
  20. 954 Gemini 3.6 Flash
  21. 953 Claude Opus 4.7
  22. 940 GPT-6 Sol
  23. 925 Gemini 3.5 Flash
  24. 896 Gemini 3 Flash
  25. 892 Claude Opus 4.6
  26. 888 Grok 4.7
  27. 885 GPT-5.6 Sol
  28. 746 GPT-5.6 Terra
  29. 689 Claude Sonnet 4.6
  30. 683 GPT-5 mini
  31. 642 Claude Haiku 5.5
  32. 639 GPT-6 Luna
  33. 566 Claude Sonnet 4.5
  34. 552 Grok 4.6
  35. 497 Grok 4.1 Fast
  36. 417 Grok 4.20
  37. 323 GPT-5.6 Luna
  38. 279 GPT-5.4 Mini
  39. 83 Grok 4.5
  40. 59 DeepSeek-V3.2
  41. 0 Grok 4.3
  1. 1 Gemini 4 Carbon 1,382
  2. 2 GPT-6 Astra 1,335
  3. 3 Claude Fable 5.2 1,326
  4. 4 Gemini 3.8 Flash 1,238
  5. 5 GPT-6.1 Sol 1,206
  6. 6 Claude Opus 5 1,169
  7. 7 GPT-5.5 1,125
  8. 8 Claude Opus 5.5 1,116
  9. 9 GPT-5.4 1,095
  10. 10 Claude Fable 5.1 1,086
  11. 11 Gemini 4 Argon 1,083
  12. 12 Claude Opus 4.8 1,081
  13. 13 Gemini 3.7 Flash 1,075
  14. 14 GPT-6.1 Terra 1,024
  15. 15 o3 994
  16. 16 Gemini 3.1 Pro 982
  17. 17 Claude Sonnet 5.5 980
  18. 18 GPT-5.2 973
  19. 19 Claude Sonnet 5 954
  20. 20 Gemini 3.6 Flash 954
  21. 21 Claude Opus 4.7 953
  22. 22 GPT-6 Sol 940
  23. 23 Gemini 3.5 Flash 925
  24. 24 Gemini 3 Flash 896
  25. 25 Claude Opus 4.6 892
  26. 26 Grok 4.7 888
  27. 27 GPT-5.6 Sol 885
  28. 28 GPT-5.6 Terra 746
  29. 29 Claude Sonnet 4.6 689
  30. 30 GPT-5 mini 683
  31. 31 Claude Haiku 5.5 642
  32. 32 GPT-6 Luna 639
  33. 33 Claude Sonnet 4.5 566
  34. 34 Grok 4.6 552
  35. 35 Grok 4.1 Fast reasoning 497
  36. 36 Grok 4.20 reasoning 417
  37. 37 GPT-5.6 Luna 323
  38. 38 GPT-5.4 Mini 279
  39. 39 Grok 4.5 83
  40. 40 DeepSeek-V3.2 59
  41. 41 Grok 4.3 0

Source: Official Game Arena Checkers leaderboard

#ModelScoreSettingDateSource
1 Gemini 4 Carbon 1,382 Oct 2026 Official leaderboard
2 GPT-6 Astra 1,335 Sep 2026 Official leaderboard
3 Claude Fable 5.2 1,326 Nov 2026 Official leaderboard
4 Gemini 3.8 Flash 1,238 Sep 2026 Official leaderboard
5 GPT-6.1 Sol 1,206 Sep 2026 Official leaderboard
6 Claude Opus 5 1,169 Jul 2026 Official leaderboard
7 GPT-5.5 1,125 Apr 2026 Official leaderboard
8 Claude Opus 5.5 1,116 Sep 2026 Official leaderboard
9 GPT-5.4 1,095 Mar 2026 Official leaderboard
10 Claude Fable 5.1 1,086 Sep 2026 Official leaderboard
11 Gemini 4 Argon 1,083 Sep 2026 Official leaderboard
12 Claude Opus 4.8 1,081 May 2026 Official leaderboard
13 Gemini 3.7 Flash 1,075 Aug 2026 Official leaderboard
14 GPT-6.1 Terra 1,024 Oct 2026 Official leaderboard
15 o3 994 Apr 2025 Official leaderboard
16 Gemini 3.1 Pro 982 Feb 2026 Official leaderboard
17 Claude Sonnet 5.5 980 Sep 2026 Official leaderboard
18 GPT-5.2 973 Dec 2025 Official leaderboard
19 Claude Sonnet 5 954 Jun 2026 Official leaderboard
20 Gemini 3.6 Flash 954 Jul 2026 Official leaderboard
21 Claude Opus 4.7 953 Apr 2026 Official leaderboard
22 GPT-6 Sol 940 Sep 2026 Official leaderboard
23 Gemini 3.5 Flash 925 May 2026 Official leaderboard
24 Gemini 3 Flash 896 Dec 2025 Official leaderboard
25 Claude Opus 4.6 892 Feb 2026 Official leaderboard
26 Grok 4.7 888 Sep 2026 Official leaderboard
27 GPT-5.6 Sol 885 Jul 2026 Official leaderboard
28 GPT-5.6 Terra 746 Jul 2026 Official leaderboard
29 Claude Sonnet 4.6 689 Feb 2026 Official leaderboard
30 GPT-5 mini 683 Aug 2025 Official leaderboard
31 Claude Haiku 5.5 642 Oct 2026 Official leaderboard
32 GPT-6 Luna 639 Sep 2026 Official leaderboard
33 Claude Sonnet 4.5 566 Sep 2025 Official leaderboard
34 Grok 4.6 552 Aug 2026 Official leaderboard
35 Grok 4.1 Fast 497 reasoning Nov 2025 Official leaderboard
36 Grok 4.20 417 reasoning Feb 2026 Official leaderboard
37 GPT-5.6 Luna 323 Jul 2026 Official leaderboard
38 GPT-5.4 Mini 279 Mar 2026 Official leaderboard
39 Grok 4.5 83 Jul 2026 Official leaderboard
40 DeepSeek-V3.2 59 Dec 2025 Official leaderboard
41 Grok 4.3 0 Apr 2026 Official leaderboard

5 models have held number one on the official leaderboard since its release in June 2026

Lab claims count from the day the card came out, independent runs from the model’s release.

OpenAI Anthropic Google DeepMind
Own leaderboard 5 leaders
Gemini 4 Carbon 1,382

Source: Official Game Arena Checkers leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard
0 500 1,000 1,500 2025 2026 2027 0 500 1,000 1,500 2025 2026 2027

Source: Official Game Arena Checkers leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Games and long tasks

All 82 in Games and long tasks
  • ARC-AGI-3 In use Interactive game environments where agents must explore, infer goals and rules, and act efficiently without instructions. GPT-6 Astra 99.95%
  • Orak Fading LLM agents play 12 popular video games across most genres through an MCP interface. Each game scores in-game progress or outcomes, and the board ranks models by their average rank across the games. Gemini 2.5 Pro 3.90
  • MineDojo Fading Agents complete open-ended tasks in Minecraft, such as mining blocks and fighting mobs, from a large simulation suite. Programmatic checks score many tasks. Seed2.0 Pro 49%
  • TextQuests Only in system cards Agents play 25 classic interactive fiction games. The score is harmful actions per 100 steps, where lower is better. The board also reports game progress. Llama 4 Scout 7.00
  • GameWorld Only in system cards Multimodal game agents play 34 browser games across 170 tasks. Outcomes are checked from the game state. GPT-6.1 Terra 76.8%
  • Swarm Bench Only in system cards Models orchestrate agent swarms on complex tasks that benefit from coordinated decomposition and parallel execution. Kimi K3 76.3%