Benchmark Almanac

Game Arena 3 Check Chess (Generality)

o3 27.0 leads since April 2025
Not in model cards

Models play both colors. A player can win by checking the opposing king three times. A double check counts once.

Generality ranks 15 configurations of 13 models across 210 games. OpenAI o3 with high reasoning won 27 of 28 games.

Made by
Generality, Inc.
Kind
Multi-agent games
Level
School
Scored
Ranked by matches
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Actions
Interaction
Multi-agent
Environment
Three-check chess
Items from
Live play or votes
Refresh
Live
Run by
A host
In views
Games, Agents, Kept fresh
Strongest model tested
GPT-5 #71 on the capability index
Last model added
September 2025
Scores refreshed
November 20th, 2026
13
Models scored on its own leaderboard

o3 leads with 27.0 among models released by September 2025

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in September 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 o3 reasoning high 27.0
  2. 2 GPT-5 reasoning high 26.0
  3. 3 Qwen3-235B-A22B-Thinking Jul 2025, reasoning medium, thinking enabled 18.0
  4. 4 Gemini 2.5 Pro Jun 2025, thinking budget -1 16.0
  5. 5 GPT-5 mini reasoning high 16.0
  6. 6 DeepSeek-V3.1 reasoning medium, thinking enabled 16.0
  7. 7 Claude Sonnet 4 thinking budget 31000 14.0
  8. 8 Claude Opus 4.1 thinking budget 31000 13.0
  9. 9 GPT-4o Aug 2024, reasoning non-reasoning model 9.00
  10. 10 DeepSeek-R1 May 2025, reasoning medium, thinking enabled 9.00
  11. 11 Llama 3.3 70B thinking non-reasoning model 3.00
  12. 12 GLM-4.5 reasoning medium, thinking enabled 2.00
  13. 13 Kimi K2 Sep 2025, thinking non-reasoning model 2.00

Source: Official Game Arena 3 Check Chess (Generality) leaderboard

#ModelScoreSettingDateSource
1 o3 27.0 reasoning high Apr 2025 Official leaderboard
2 GPT-5 26.0 reasoning high Aug 2025 Official leaderboard
3 Qwen3-235B-A22B-Thinking Jul 2025 18.0 reasoning medium, thinking enabled Jul 2025 Official leaderboard
4 Gemini 2.5 Pro Jun 2025 16.0 thinking budget -1 Jun 2025 Official leaderboard
5 GPT-5 mini 16.0 reasoning high Aug 2025 Official leaderboard
6 DeepSeek-V3.1 16.0 reasoning medium, thinking enabled Aug 2025 Official leaderboard
7 Claude Sonnet 4 14.0 thinking budget 31000 May 2025 Official leaderboard
8 Claude Opus 4.1 13.0 thinking budget 31000 Aug 2025 Official leaderboard
9 GPT-4o Aug 2024 9.00 reasoning non-reasoning model Aug 2024 Official leaderboard
10 DeepSeek-R1 May 2025 9.00 reasoning medium, thinking enabled May 2025 Official leaderboard
11 Llama 3.3 70B 3.00 thinking non-reasoning model Dec 2024 Official leaderboard
12 GLM-4.5 2.00 reasoning medium, thinking enabled Aug 2025 Official leaderboard
13 Kimi K2 Sep 2025 2.00 thinking non-reasoning model Sep 2025 Official leaderboard

2 models have held number one on the official leaderboard since 2023

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

OpenAI
Own leaderboard Not tested since Sep 2025
o3 27.0

Source: Official Game Arena 3 Check Chess (Generality) leaderboard

Scores over time

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

Plot with other benchmarks
On the benchmark’s own leaderboard
0.00 10.0 20.0 30.0 2025 0.00 10.0 20.0 30.0 2025

Source: Official Game Arena 3 Check Chess (Generality) leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Games and long tasks

All 261 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-6.2 Astra 100%
  • 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%

Data audit

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 14 months ago, in September 2025, for Kimi K2. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (13 entries).
  • Pass Models scored 13 models we track.
  • Warning Newest score The newest came 14 months ago, in September 2025, for Kimi K2.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5 (#71).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on game-arena.ai: 15 rows for models we track at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check