Benchmark Almanac

Crest Arcade Sudoku

Claude Opus 5.5 100% leads since September 2026
Not in model cards

Models fill complete Sudoku grids and report confidence. The score averages the share solved across seven equally weighted difficulty tiers.

Glimmer Index publishes scores for 18 distinct models. Of those, 12 completed the required set.

Made by
Glimmer Index
Kind
Puzzles and chess
Level
Everyday
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
GPT-6.2 Astra #1 on the capability index
Last model added
November 2026
Scores refreshed
November 20th, 2026
18
Entries on its own leaderboard, 17 models we track

3 models share the top score of 100% 3 models share the top score of 100%

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 Claude Opus 5.5 medium 100%
  2. 2 GPT-6.2 Astra medium 100%
  3. 3 Gemini 4 Neon medium · Provisional 100%
  4. 4 Gemini 4 Hydrogen medium 95.2%
  5. 5 Claude Fable 5.1 medium 92.9%
  6. 6 GPT-6 Sol medium 90.5%
  7. 7 Claude Sonnet 5.5 medium 88.1%
  8. 8 GPT-6.1 Sol medium 83.3%
  9. 9 GPT-6 Astra medium 76.2%
  10. 10 Qwen3.8-Flash-Next low 47.6%
  11. 11 Claude Haiku 5.5 medium 31%
  12. 12 gpt-oss-20b low · Provisional 28.6%
  13. 13 GPT-6.1 Terra medium 7.10%
  14. 14 Gemma 3 4B none · Provisional 0%
  15. 15 Qwen3.5-9B none · Provisional 0%
  16. 16 Qwen 3.8 27B none · Provisional 0%
  17. 17 GPT-6 Luna medium 0%
  18. 18 Ministral 3 8B none · Provisional 0%

Source: Official Crest Arcade Sudoku leaderboard

#EntryScore ±SettingReleasedRun onSource
1 Claude Opus 5.5 100% 91.3 to 100 medium Sep 2026 Oct 2026 Official leaderboard
2 GPT-6.2 Astra 100% 91.3 to 100 medium Nov 2026 Nov 2026 Official leaderboard
3 Gemini 4 Neon 100% 84 to 100 medium · Provisional Nov 2026 Nov 2026 Official leaderboard
4 Gemini 4 Hydrogen 95.2% 85.7 to 100 medium Nov 2026 Nov 2026 Official leaderboard
5 Claude Fable 5.1 92.9% 83.1 to 100 medium Sep 2026 Oct 2026 Official leaderboard
6 GPT-6 Sol 90.5% 80.5 to 100 medium Sep 2026 Oct 2026 Official leaderboard
7 Claude Sonnet 5.5 88.1% ±10.5 medium Sep 2026 Oct 2026 Official leaderboard
8 GPT-6.1 Sol 83.3% ±11.1 medium Sep 2026 Oct 2026 Official leaderboard
9 GPT-6 Astra 76.2% ±11.8 medium Sep 2026 Oct 2026 Official leaderboard
10 Qwen3.8-Flash-Next 47.6% ±12.2 low Aug 2026 Oct 2026 Official leaderboard
11 Claude Haiku 5.5 31% ±12.3 medium Oct 2026 Oct 2026 Official leaderboard
12 gpt-oss-20b 28.6% ±20.2 low · Provisional Aug 2025 Oct 2026 Official leaderboard
13 GPT-6.1 Terra 7.10% 0 to 16.6 medium Oct 2026 Oct 2026 Official leaderboard
14 Gemma 3 4B 0% 0 to 13.3 none · Provisional Mar 2025 Oct 2026 Official leaderboard
15 Qwen3.5-9B 0% 0 to 13 none · Provisional Feb 2026 Oct 2026 Official leaderboard
16 Qwen 3.8 27B 0% 0 to 13.3 none · Provisional Aug 2026 Oct 2026 Official leaderboard
17 GPT-6 Luna 0% 0 to 8.7 medium Sep 2026 Oct 2026 Official leaderboard
18 Ministral 3 8B open weights 0% 0 to 13.3 none · Provisional Oct 2026 Official leaderboard
#ModelScore ±SettingReleasedRun onSource
1 Claude Opus 5.5 100% 91.3 to 100 medium Sep 2026 Oct 2026 Official leaderboard
2 GPT-6.2 Astra 100% 91.3 to 100 medium Nov 2026 Nov 2026 Official leaderboard
3 Gemini 4 Neon 100% 84 to 100 medium · Provisional Nov 2026 Nov 2026 Official leaderboard
4 Gemini 4 Hydrogen 95.2% 85.7 to 100 medium Nov 2026 Nov 2026 Official leaderboard
5 Claude Fable 5.1 92.9% 83.1 to 100 medium Sep 2026 Oct 2026 Official leaderboard
6 GPT-6 Sol 90.5% 80.5 to 100 medium Sep 2026 Oct 2026 Official leaderboard
7 Claude Sonnet 5.5 88.1% ±10.5 medium Sep 2026 Oct 2026 Official leaderboard
8 GPT-6.1 Sol 83.3% ±11.1 medium Sep 2026 Oct 2026 Official leaderboard
9 GPT-6 Astra 76.2% ±11.8 medium Sep 2026 Oct 2026 Official leaderboard
10 Qwen3.8-Flash-Next 47.6% ±12.2 low Aug 2026 Oct 2026 Official leaderboard
11 Claude Haiku 5.5 31% ±12.3 medium Oct 2026 Oct 2026 Official leaderboard
12 gpt-oss-20b 28.6% ±20.2 low · Provisional Aug 2025 Oct 2026 Official leaderboard
13 GPT-6.1 Terra 7.10% 0 to 16.6 medium Oct 2026 Oct 2026 Official leaderboard
14 Gemma 3 4B 0% 0 to 13.3 none · Provisional Mar 2025 Oct 2026 Official leaderboard
15 Qwen3.5-9B 0% 0 to 13 none · Provisional Feb 2026 Oct 2026 Official leaderboard
16 Qwen 3.8 27B 0% 0 to 13.3 none · Provisional Aug 2026 Oct 2026 Official leaderboard
17 GPT-6 Luna 0% 0 to 8.7 medium Sep 2026 Oct 2026 Official leaderboard

Rankings by part

We read 7 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Cozy, Gentle and Steady. Claude Fable 5.1 leads 3 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Cozy

17 models we track have a score here.

  1. 1 Claude Fable 5.1 ↑4 100%
  2. 2 Claude Opus 5.5 ↓1 100%
  3. 3 GPT-6 Sol ↑3 100%
  4. 4 GPT-6.1 Sol ↑4 100%
  5. 5 Gemini 4 Hydrogen ↓1 100%
  6. 6 GPT-6.2 Astra ↓4 100%
  7. 7 Gemini 4 Neon ↓4 100%
  8. 8 GPT-6 Astra ↑1 83.3%
  9. 9 Claude Sonnet 5.5 ↓2 83.3%
  10. 10 Qwen3.8-Flash-Next 66.7%

The best score, 100%, costs $0.20 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 Anthropic Google DeepMind Cost frontier One model at each effort level
0% 20% 40% 60% 80% 100% $0.001 $0.01 $0.10 $1.00 Cost per task, log scale GPT-6.2 Astra, medium Gemini 4 Hydrogen, medium GPT-6 Sol, medium GPT-6.1 Sol, medium Claude Haiku 5.5, medium GPT-6.1 Terra, medium GPT-6 Luna, medium 0% 20% 40% 60% 80% 100% $0.001 $0.01 $0.10 $1.00 Cost per task, log scale GPT-6.2 Astra, medium GPT-6 Sol, medium

Source: Official Crest Arcade Sudoku leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Opus 5.5 medium 100% $0.24 11,827
GPT-6.2 Astra medium 100% $0.20 3,538
Gemini 4 Neon medium · Provisional 100% $0.30 19,454
Gemini 4 Hydrogen medium 95.2% $0.19 24,155
Claude Fable 5.1 medium 92.9% $0.89 16,999
GPT-6 Sol medium 90.5% $0.15 12,925
Claude Sonnet 5.5 medium 88.1% $0.14 13,462
GPT-6.1 Sol medium 83.3% $0.11 8,723
GPT-6 Astra medium 76.2% $0.45 6,987
Qwen3.8-Flash-Next low 47.6% 65,074
Claude Haiku 5.5 medium 31% $0.02 30,157
gpt-oss-20b low · Provisional 28.6% 29,731
GPT-6.1 Terra medium 7.10% $0.01 1,849
Gemma 3 4B none · Provisional 0% 166
Qwen3.5-9B none · Provisional 0% 2,362
Qwen 3.8 27B none · Provisional 0% 11,673
GPT-6 Luna medium 0% $0.002 1,779

The best score, 100%, costs $0.20 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 Anthropic Google DeepMind Cost frontier One model at each effort level
0% 20% 40% 60% 80% 100% $0.001 $0.01 $0.10 $1.00 Cost per task, log scale GPT-6.2 Astra, medium Gemini 4 Hydrogen, medium GPT-6 Sol, medium GPT-6.1 Sol, medium Claude Haiku 5.5, medium GPT-6.1 Terra, medium GPT-6 Luna, medium 0% 20% 40% 60% 80% 100% $0.001 $0.01 $0.10 $1.00 Cost per task, log scale GPT-6.2 Astra, medium GPT-6 Sol, medium

Source: Official Crest Arcade Sudoku leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Opus 5.5 medium 100% $0.24 11,827
GPT-6.2 Astra medium 100% $0.20 3,538
Gemini 4 Neon medium · Provisional 100% $0.30 19,454
Gemini 4 Hydrogen medium 95.2% $0.19 24,155
Claude Fable 5.1 medium 92.9% $0.89 16,999
GPT-6 Sol medium 90.5% $0.15 12,925
Claude Sonnet 5.5 medium 88.1% $0.14 13,462
GPT-6.1 Sol medium 83.3% $0.11 8,723
GPT-6 Astra medium 76.2% $0.45 6,987
Qwen3.8-Flash-Next low 47.6% 65,074
Claude Haiku 5.5 medium 31% $0.02 30,157
gpt-oss-20b low · Provisional 28.6% 29,731
GPT-6.1 Terra medium 7.10% $0.01 1,849
Gemma 3 4B none · Provisional 0% 166
Qwen3.5-9B none · Provisional 0% 2,362
Qwen 3.8 27B none · Provisional 0% 11,673
GPT-6 Luna medium 0% $0.002 1,779

5 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 Qwen Anthropic
Own leaderboard Passed 85% in Sep 2026, by Claude Fable 5.1

Source: Official Crest Arcade Sudoku leaderboard

Claude Fable 5.1 on the official leaderboard passed 85% in September 2026

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 Best so far
0% 20% 40% 60% 80% 100% 2025 2026 2027 Near the ceiling (85%) Claude Opus 5.5 Claude Fable 5.1 Qwen3.8-Flash-Next gpt-oss-20b Gemma 3 4B 0% 20% 40% 60% 80% 100% 2025 2026 2027 Near the ceiling (85%) Claude Opus 5.5 Claude Fable 5.1

Source: Official Crest Arcade Sudoku 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 (18 entries).
  • Pass Models scored 17 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 7 of the top 10 have a score, led by GPT-6.2 Astra (#1).
  • Note Headroom Saturated. The best is 100%, past 85% since September 2026.
Leaderboard import 3 pass
  • Pass Importer We read the board on glimmerindex.com: 17 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