Benchmark Almanac

Crest Arcade Nations Royale

gpt-oss-20b 100.0 leads since August 2025
Not in model cards

Models manage one nation against 47 computer opponents on a simulated map. Wins earn full credit and unfinished games earn credit for territory held.

Glimmer Index publishes scores for 17 distinct models. Of those, 10 completed the required set.

Made by
Glimmer Index
Kind
Video games
Level
Everyday
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Actions
Interaction
Multi-agent
Environment
Nations Royale
Items from
Generated
Refresh
One fixed set
Run by
Anyone
In views
Games, Agents
Strongest model tested
GPT-6.2 Astra #1 on the capability index
Last model added
November 2026
Scores refreshed
November 20th, 2026
17
Entries on its own leaderboard, 16 models we track

2 models share the top score of 100.0 2 models share the top score of 100.0

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.

  1. 1 gpt-oss-20b low · Provisional 100.0
  2. 2 Gemini 4 Argon medium · Provisional 100.0
  3. 3 GPT-6 Astra medium 91.7
  4. 4 GPT-6.1 Sol medium 91.7
  5. 5 GPT-6.1 Terra medium 91.7
  6. 6 Gemini 4 Hydrogen medium 91.7
  7. 7 Gemini 4 Neon medium · Provisional 87.5
  8. 8 Qwen3.8-Flash-Next low 83.3
  9. 9 GPT-6 Sol medium 83.3
  10. 10 Claude Sonnet 5.5 medium 83.3
  11. 11 GPT-6.2 Astra medium 83.3
  12. 12 GPT-6 Luna medium 78.6
  13. 13 Claude Haiku 5.5 medium 78.3
  14. 14 Qwen3.5-9B none · Provisional 27.1
  15. 15 Qwen 3.8 27B none · Provisional 25.0
  16. 16 Ministral 3 8B none · Provisional 25.0
  17. 17 Gemma 3 4B none · Provisional 1.90

Source: Official Crest Arcade Nations Royale leaderboard

#EntryScore ±SettingReleasedRun onSource
1 gpt-oss-20b 100.0 69.2 to 100 low · Provisional Aug 2025 Nov 2026 Official leaderboard
2 Gemini 4 Argon 100.0 62.3 to 100 medium · Provisional Sep 2026 Nov 2026 Official leaderboard
3 GPT-6 Astra 91.7 73.1 to 100 medium Sep 2026 Oct 2026 Official leaderboard
4 GPT-6.1 Sol 91.7 73.1 to 100 medium Sep 2026 Oct 2026 Official leaderboard
5 GPT-6.1 Terra 91.7 73.1 to 100 medium Oct 2026 Oct 2026 Official leaderboard
6 Gemini 4 Hydrogen 91.7 73.1 to 100 medium Nov 2026 Nov 2026 Official leaderboard
7 Gemini 4 Neon 87.5 65.4 to 100 medium · Provisional Nov 2026 Nov 2026 Official leaderboard
8 Qwen3.8-Flash-Next 83.3 64.7 to 100 low Aug 2026 Nov 2026 Official leaderboard
9 GPT-6 Sol 83.3 64.7 to 100 medium Sep 2026 Oct 2026 Official leaderboard
10 Claude Sonnet 5.5 83.3 64.7 to 100 medium Sep 2026 Nov 2026 Official leaderboard
11 GPT-6.2 Astra 83.3 64.7 to 100 medium Nov 2026 Nov 2026 Official leaderboard
12 GPT-6 Luna 78.6 ±18.1 medium Sep 2026 Oct 2026 Official leaderboard
13 Claude Haiku 5.5 78.3 ±18.1 medium Oct 2026 Nov 2026 Official leaderboard
14 Qwen3.5-9B 27.1 ±26.8 none · Provisional Feb 2026 Nov 2026 Official leaderboard
15 Qwen 3.8 27B 25.0 ±25.2 none · Provisional Aug 2026 Nov 2026 Official leaderboard
16 Ministral 3 8B open weights 25.0 0 to 51.7 none · Provisional Nov 2026 Official leaderboard
17 Gemma 3 4B 1.90 0 to 28.7 none · Provisional Mar 2025 Nov 2026 Official leaderboard
#ModelScore ±SettingReleasedRun onSource
1 gpt-oss-20b 100.0 69.2 to 100 low · Provisional Aug 2025 Nov 2026 Official leaderboard
2 Gemini 4 Argon 100.0 62.3 to 100 medium · Provisional Sep 2026 Nov 2026 Official leaderboard
3 GPT-6 Astra 91.7 73.1 to 100 medium Sep 2026 Oct 2026 Official leaderboard
4 GPT-6.1 Sol 91.7 73.1 to 100 medium Sep 2026 Oct 2026 Official leaderboard
5 GPT-6.1 Terra 91.7 73.1 to 100 medium Oct 2026 Oct 2026 Official leaderboard
6 Gemini 4 Hydrogen 91.7 73.1 to 100 medium Nov 2026 Nov 2026 Official leaderboard
7 Gemini 4 Neon 87.5 65.4 to 100 medium · Provisional Nov 2026 Nov 2026 Official leaderboard
8 Qwen3.8-Flash-Next 83.3 64.7 to 100 low Aug 2026 Nov 2026 Official leaderboard
9 GPT-6 Sol 83.3 64.7 to 100 medium Sep 2026 Oct 2026 Official leaderboard
10 Claude Sonnet 5.5 83.3 64.7 to 100 medium Sep 2026 Nov 2026 Official leaderboard
11 GPT-6.2 Astra 83.3 64.7 to 100 medium Nov 2026 Nov 2026 Official leaderboard
12 GPT-6 Luna 78.6 ±18.1 medium Sep 2026 Oct 2026 Official leaderboard
13 Claude Haiku 5.5 78.3 ±18.1 medium Oct 2026 Nov 2026 Official leaderboard
14 Qwen3.5-9B 27.1 ±26.8 none · Provisional Feb 2026 Nov 2026 Official leaderboard
15 Qwen 3.8 27B 25.0 ±25.2 none · Provisional Aug 2026 Nov 2026 Official leaderboard
16 Gemma 3 4B 1.90 0 to 28.7 none · Provisional Mar 2025 Nov 2026 Official leaderboard

Rankings by part

We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Easy, Medium and Hard. gpt-oss-20b leads 3 of them, and GPT-6 Astra leads the other one. Arrows show how far a model moved from its overall rank among the models we track.

Easy

16 models we track have a score here.

  1. 1 gpt-oss-20b 100.0
  2. 2 Qwen3.5-9B ↑12 100.0
  3. 3 Qwen 3.8 27B ↑12 100.0
  4. 4 Qwen3.8-Flash-Next ↑4 100.0
  5. 5 GPT-6 Astra ↓2 100.0
  6. 6 GPT-6 Luna ↑6 100.0
  7. 7 GPT-6 Sol ↑2 100.0
  8. 8 Claude Sonnet 5.5 ↑2 100.0
  9. 9 GPT-6.1 Sol ↓5 100.0
  10. 10 Gemini 4 Argon ↓8 100.0

The best score, 100.0, costs $0.49 a task

Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

OpenAI Google DeepMind Cost frontier One model at each effort level
0.00 20.0 40.0 60.0 80.0 100.0 $0.01 $0.10 $1.00 $10 Cost per task, log scale Gemini 4 Argon, medium · Provisional GPT-6.1 Terra, medium GPT-6 Luna, medium 0.00 20.0 40.0 60.0 80.0 100.0 $0.01 $0.10 $1.00 $10 Cost per task, log scale Gemini 4 Argon, medium · Provisional GPT-6.1 Terra, medium

Source: Official Crest Arcade Nations Royale leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
gpt-oss-20b low · Provisional 100.0
Gemini 4 Argon medium · Provisional 100.0 $0.49
GPT-6 Astra medium 91.7 $2.41
GPT-6.1 Sol medium 91.7 $0.53
GPT-6.1 Terra medium 91.7 $0.29
Gemini 4 Hydrogen medium 91.7 $1.43
Gemini 4 Neon medium · Provisional 87.5 $2.91
Qwen3.8-Flash-Next low 83.3
GPT-6 Sol medium 83.3 $0.89
Claude Sonnet 5.5 medium 83.3 $1.32
GPT-6.2 Astra medium 83.3 $1.06
GPT-6 Luna medium 78.6 $0.05
Claude Haiku 5.5 medium 78.3 $0.14
Qwen3.5-9B none · Provisional 27.1
Qwen 3.8 27B none · Provisional 25.0
Gemma 3 4B none · Provisional 1.90

The best score, 100.0, costs $0.49 a task

Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

OpenAI Google DeepMind Cost frontier One model at each effort level
0.00 20.0 40.0 60.0 80.0 100.0 $0.01 $0.10 $1.00 $10 Cost per task, log scale Gemini 4 Argon, medium · Provisional GPT-6.1 Terra, medium GPT-6 Luna, medium 0.00 20.0 40.0 60.0 80.0 100.0 $0.01 $0.10 $1.00 $10 Cost per task, log scale Gemini 4 Argon, medium · Provisional GPT-6.1 Terra, medium

Source: Official Crest Arcade Nations Royale leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
gpt-oss-20b low · Provisional 100.0
Gemini 4 Argon medium · Provisional 100.0 $0.49
GPT-6 Astra medium 91.7 $2.41
GPT-6.1 Sol medium 91.7 $0.53
GPT-6.1 Terra medium 91.7 $0.29
Gemini 4 Hydrogen medium 91.7 $1.43
Gemini 4 Neon medium · Provisional 87.5 $2.91
Qwen3.8-Flash-Next low 83.3
GPT-6 Sol medium 83.3 $0.89
Claude Sonnet 5.5 medium 83.3 $1.32
GPT-6.2 Astra medium 83.3 $1.06
GPT-6 Luna medium 78.6 $0.05
Claude Haiku 5.5 medium 78.3 $0.14
Qwen3.5-9B none · Provisional 27.1
Qwen 3.8 27B none · Provisional 25.0
Gemma 3 4B none · Provisional 1.90

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.

Google DeepMind OpenAI
Own leaderboard 2 leaders
gpt-oss-20b 100.0

Source: Official Crest Arcade Nations Royale 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 20.0 40.0 60.0 80.0 100.0 2025 2026 2027 0.00 20.0 40.0 60.0 80.0 100.0 2025 2026 2027

Source: Official Crest Arcade Nations Royale 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

All 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (17 entries).
  • Pass Models scored 16 models we track, plus 1 other entry on its leaderboard.
  • Pass Newest score The newest is from November 2026, for Gemini 4 Neon.
  • Pass Frontier models 6 of the top 10 have a score, led by GPT-6.2 Astra (#1).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on glimmerindex.com: 16 rows for models we track and 1 other entry 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