Benchmark Almanac

Game Arena Coin Game

GPT-5.5 514 leads since Jul 2026 Not in model cards

Two language models move on a grid and pick up coins. Each should collect the coins that score for it and avoid the rest.

Kaggle published the leaderboard on 2026-07-08 with 35 models on 2026-11-01. Gameplay logs are a public Kaggle dataset. Part of the Kaggle Game Arena suite.

Made by Kaggle and Google DeepMind Released Jul 2026 Multi-agent games All tasks public Last model added Oct 2026 Scores refreshed 2 Nov 2026
Level
Everyday Most adults, without training
Scored
Ranked by matches Models play against each other, and the results become ratings.
Answers
Agent task Many steps in a terminal, browser, desktop, app, game or robot.
Kaggle/kaggle-environments 459 stars Dataset Leaderboard kaggle.com
35
Models scored on its own leaderboard

GPT-5.5 leads with 514

Best score per model from the benchmark's own leaderboard, read on 2 Nov 2026. Where a model ran with several settings or harnesses, we keep its best.

35 models. Scroll sideways for the rest.

  1. 514 GPT-5.5
  2. 488 Gemini 3.8 Flash
  3. 484 Gemini 3.7 Flash
  4. 472 Gemini 3 Flash
  5. 460 Gemini 4 Argon
  6. 438 GPT-6 Sol
  7. 438 Gemini 3.6 Flash
  8. 421 GPT-6 Astra
  9. 420 Gemini 3.1 Pro
  10. 418 GPT-6.1 Sol
  11. 417 GPT-5.4
  12. 415 GPT-5.6 Sol
  13. 401 Claude Opus 5
  14. 382 Gemini 3.5 Flash
  15. 376 Grok 4.7
  16. 368 Grok 4.6
  17. 367 Grok 4.5
  18. 355 Claude Sonnet 4.5
  19. 350 Claude Opus 4.8
  20. 313 Claude Opus 4.7
  21. 310 Claude Opus 4.6
  22. 287 Claude Sonnet 4.6
  23. 279 Grok 4.3
  24. 248 Grok 4.20
  25. 246 Claude Opus 5.5
  26. 241 GPT-5.2
  27. 236 Claude Sonnet 5
  28. 232 Claude Haiku 5.5
  29. 223 GPT-5 mini
  30. 200 GPT-5.6 Terra
  31. 196 Claude Sonnet 5.5
  32. 174 GPT-6 Luna
  33. 147 o3
  34. 73 GPT-5.4 Mini
  35. 0 GPT-5.6 Luna
  1. 1 GPT-5.5 514
  2. 2 Gemini 3.8 Flash 488
  3. 3 Gemini 3.7 Flash 484
  4. 4 Gemini 3 Flash 472
  5. 5 Gemini 4 Argon 460
  6. 6 GPT-6 Sol 438
  7. 7 Gemini 3.6 Flash 438
  8. 8 GPT-6 Astra 421
  9. 9 Gemini 3.1 Pro 420
  10. 10 GPT-6.1 Sol 418
  11. 11 GPT-5.4 417
  12. 12 GPT-5.6 Sol 415
  13. 13 Claude Opus 5 401
  14. 14 Gemini 3.5 Flash 382
  15. 15 Grok 4.7 376
  16. 16 Grok 4.6 368
  17. 17 Grok 4.5 367
  18. 18 Claude Sonnet 4.5 355
  19. 19 Claude Opus 4.8 350
  20. 20 Claude Opus 4.7 313
  21. 21 Claude Opus 4.6 310
  22. 22 Claude Sonnet 4.6 287
  23. 23 Grok 4.3 279
  24. 24 Grok 4.20 reasoning 248
  25. 25 Claude Opus 5.5 246
  26. 26 GPT-5.2 241
  27. 27 Claude Sonnet 5 236
  28. 28 Claude Haiku 5.5 232
  29. 29 GPT-5 mini 223
  30. 30 GPT-5.6 Terra 200
  31. 31 Claude Sonnet 5.5 196
  32. 32 GPT-6 Luna 174
  33. 33 o3 147
  34. 34 GPT-5.4 Mini 73
  35. 35 GPT-5.6 Luna 0
#ModelScoreSettingDateSource
1 GPT-5.5 514 Apr 2026 Official leaderboard
2 Gemini 3.8 Flash 488 Sep 2026 Official leaderboard
3 Gemini 3.7 Flash 484 Aug 2026 Official leaderboard
4 Gemini 3 Flash 472 Dec 2025 Official leaderboard
5 Gemini 4 Argon 460 Sep 2026 Official leaderboard
6 GPT-6 Sol 438 Sep 2026 Official leaderboard
7 Gemini 3.6 Flash 438 Jul 2026 Official leaderboard
8 GPT-6 Astra 421 Sep 2026 Official leaderboard
9 Gemini 3.1 Pro 420 Feb 2026 Official leaderboard
10 GPT-6.1 Sol 418 Sep 2026 Official leaderboard
11 GPT-5.4 417 Mar 2026 Official leaderboard
12 GPT-5.6 Sol 415 Jul 2026 Official leaderboard
13 Claude Opus 5 401 Jul 2026 Official leaderboard
14 Gemini 3.5 Flash 382 May 2026 Official leaderboard
15 Grok 4.7 376 Sep 2026 Official leaderboard
16 Grok 4.6 368 Aug 2026 Official leaderboard
17 Grok 4.5 367 Jul 2026 Official leaderboard
18 Claude Sonnet 4.5 355 Sep 2025 Official leaderboard
19 Claude Opus 4.8 350 May 2026 Official leaderboard
20 Claude Opus 4.7 313 Apr 2026 Official leaderboard
21 Claude Opus 4.6 310 Feb 2026 Official leaderboard
22 Claude Sonnet 4.6 287 Feb 2026 Official leaderboard
23 Grok 4.3 279 Apr 2026 Official leaderboard
24 Grok 4.20 248 reasoning Feb 2026 Official leaderboard
25 Claude Opus 5.5 246 Sep 2026 Official leaderboard
26 GPT-5.2 241 Dec 2025 Official leaderboard
27 Claude Sonnet 5 236 Jun 2026 Official leaderboard
28 Claude Haiku 5.5 232 Oct 2026 Official leaderboard
29 GPT-5 mini 223 Aug 2025 Official leaderboard
30 GPT-5.6 Terra 200 Jul 2026 Official leaderboard
31 Claude Sonnet 5.5 196 Sep 2026 Official leaderboard
32 GPT-6 Luna 174 Sep 2026 Official leaderboard
33 o3 147 Apr 2025 Official leaderboard
34 GPT-5.4 Mini 73 Mar 2026 Official leaderboard
35 GPT-5.6 Luna 0 Jul 2026 Official leaderboard

Number one over time

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

OpenAI
Own leaderboard 1 leader
GPT-5.5 514

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
0 200 400 600 2025 2026 0 200 400 600 2025 2026

No lab model card reports it yet

We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.

More in Games and long tasks

All 68 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 100%
  • 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%
  • Orak Fading LLM agents play 12 popular video games across most genres through an MCP interface. Scores come from in-game progress or outcomes per title. Nemotron 3 Nano 4B 22.9
  • GameWorld Only in system cards Multimodal game agents play 34 browser games across 170 tasks. Outcomes are checked from the game state. Gemini 3 Flash 41.9%
  • 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%
  • TextQuests Only in system cards Agents play 25 classic interactive fiction games. Measures game progress and harmful actions per 100 steps. Muse Spark 2.40