Benchmark Almanac

The Aggregate Inversion and Execution Ladder

GPT-5.5 100% leads since April 2026
Not in model cards

Models predict the output of a short program or reconstruct the input to a digit transformation. The scorer checks exact answers across six difficulty levels.

The maker tracks 27 models on 24 fixed puzzles. Providers use their default reasoning settings, and some models answered fewer puzzles.

Kind
Puzzles and chess
Level
School
Scored
Checked automatically
Answers
Short answer
Measures
Capability
Language
English
In and out
Code, Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
A host
Strongest model tested
GPT-6.2 Astra #1 on the capability index
Last model added
November 2026
Scores refreshed
November 20th, 2026
27
Entries on its own leaderboard, 20 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. 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-5.5 unset 100%
  2. 2 Claude Fable 5.2 unset 100%
  3. 3 GPT-6.2 Astra unset 100%
  4. 4 Dots3-Note (Preview) free route, unset 100%
  5. 5 ERNIE 5.5 Flash Preview free route, unset 100%
  6. 6 Particle-1 free route, unset 100%
  7. 7 GPT-5.2 unset 91.7%
  8. 8 GLM-5.3 Mistral API, unset 91.7%
  9. 9 Ling-3.1-flash unset 91.7%
  10. 10 Nemotron 3.5 Super VL free route, unset 91.7%
  11. 11 Kimi K2.6 unset 87.5%
  12. 12 GPT-6.2 Sol unset 87.5%
  13. 13 Nemotron 3 Super free route, unset 83.3%
  14. 14 Claude Opus 4.8 unset 83.3%
  15. 15 GLM-5.3-Flash Mistral API, unset 83.3%
  16. 16 DeepSeek V4.1 Flash 1017 Mistral API, unset 83.3%
  17. 17 gpt-oss-120b unset 79.2%
  18. 18 Ling-3.0-flash-VL free route, unset 78.3%
  19. 19 MiniMax-M3.1-Flash free route, unset 75%
  20. 20 Nex N2.5 Pro free route, unset 70.8%
  21. 21 Flint free route, unset 54.5%
  22. 22 Nex N2.5 Mini free route, unset 50%
  23. 23 GPT-5.4 unset 41.7%
  24. 24 Gemma 4 26B A4B free route, unset 41.7%
  25. 25 Space Bunny Alpha stealth release, unset 26.1%
  26. 26 Gemma 4 31B IT free route, unset 25%
  27. 27 Kimi K2 Sep 2025, unset 20.8%

Source: Official The Aggregate Inversion and Execution Ladder leaderboard

#EntryScoreSettingReleasedRun onSource
1 GPT-5.5 100% unset Apr 2026 Nov 2026 Official leaderboard
2 Claude Fable 5.2 100% unset Nov 2026 Nov 2026 Official leaderboard
3 GPT-6.2 Astra 100% unset Nov 2026 Nov 2026 Official leaderboard
4 Dots3-Note (Preview) 100% free route, unset Nov 2026 Official leaderboard
5 ERNIE 5.5 Flash Preview 100% free route, unset Nov 2026 Official leaderboard
6 Particle-1 100% free route, unset Nov 2026 Official leaderboard
7 GPT-5.2 91.7% unset Dec 2025 Nov 2026 Official leaderboard
8 GLM-5.3 91.7% Mistral API, unset Aug 2026 Nov 2026 Official leaderboard
9 Ling-3.1-flash 91.7% unset Oct 2026 Nov 2026 Official leaderboard
10 Nemotron 3.5 Super VL 91.7% free route, unset Oct 2026 Nov 2026 Official leaderboard
11 Kimi K2.6 87.5% unset Apr 2026 Nov 2026 Official leaderboard
12 GPT-6.2 Sol 87.5% unset Nov 2026 Nov 2026 Official leaderboard
13 Nemotron 3 Super 83.3% free route, unset Mar 2026 Nov 2026 Official leaderboard
14 Claude Opus 4.8 83.3% unset May 2026 Nov 2026 Official leaderboard
15 GLM-5.3-Flash 83.3% Mistral API, unset Aug 2026 Nov 2026 Official leaderboard
16 DeepSeek V4.1 Flash 1017 83.3% Mistral API, unset Oct 2026 Nov 2026 Official leaderboard
17 gpt-oss-120b 79.2% unset Aug 2025 Nov 2026 Official leaderboard
18 Ling-3.0-flash-VL 78.3% free route, unset Nov 2026 Official leaderboard
19 MiniMax-M3.1-Flash 75% free route, unset Oct 2026 Nov 2026 Official leaderboard
20 Nex N2.5 Pro 70.8% free route, unset Nov 2026 Official leaderboard
21 Flint 54.5% free route, unset Nov 2026 Nov 2026 Official leaderboard
22 Nex N2.5 Mini 50% free route, unset Nov 2026 Official leaderboard
23 GPT-5.4 41.7% unset Mar 2026 Nov 2026 Official leaderboard
24 Gemma 4 26B A4B 41.7% free route, unset Apr 2026 Nov 2026 Official leaderboard
25 Space Bunny Alpha 26.1% stealth release, unset Nov 2026 Official leaderboard
26 Gemma 4 31B IT 25% free route, unset Apr 2026 Nov 2026 Official leaderboard
27 Kimi K2 Sep 2025 20.8% unset Sep 2025 Nov 2026 Official leaderboard
#ModelScoreSettingReleasedRun onSource
1 GPT-5.5 100% unset Apr 2026 Nov 2026 Official leaderboard
2 Claude Fable 5.2 100% unset Nov 2026 Nov 2026 Official leaderboard
3 GPT-6.2 Astra 100% unset Nov 2026 Nov 2026 Official leaderboard
4 GPT-5.2 91.7% unset Dec 2025 Nov 2026 Official leaderboard
5 GLM-5.3 91.7% Mistral API, unset Aug 2026 Nov 2026 Official leaderboard
6 Ling-3.1-flash 91.7% unset Oct 2026 Nov 2026 Official leaderboard
7 Nemotron 3.5 Super VL 91.7% free route, unset Oct 2026 Nov 2026 Official leaderboard
8 Kimi K2.6 87.5% unset Apr 2026 Nov 2026 Official leaderboard
9 GPT-6.2 Sol 87.5% unset Nov 2026 Nov 2026 Official leaderboard
10 Nemotron 3 Super 83.3% free route, unset Mar 2026 Nov 2026 Official leaderboard
11 Claude Opus 4.8 83.3% unset May 2026 Nov 2026 Official leaderboard
12 GLM-5.3-Flash 83.3% Mistral API, unset Aug 2026 Nov 2026 Official leaderboard
13 DeepSeek V4.1 Flash 1017 83.3% Mistral API, unset Oct 2026 Nov 2026 Official leaderboard
14 gpt-oss-120b 79.2% unset Aug 2025 Nov 2026 Official leaderboard
15 MiniMax-M3.1-Flash 75% free route, unset Oct 2026 Nov 2026 Official leaderboard
16 Flint 54.5% free route, unset Nov 2026 Nov 2026 Official leaderboard
17 GPT-5.4 41.7% unset Mar 2026 Nov 2026 Official leaderboard
18 Gemma 4 26B A4B 41.7% free route, unset Apr 2026 Nov 2026 Official leaderboard
19 Gemma 4 31B IT 25% free route, unset Apr 2026 Nov 2026 Official leaderboard
20 Kimi K2 Sep 2025 20.8% unset Sep 2025 Nov 2026 Official leaderboard

3 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 Passed 85% in Dec 2025, by GPT-5.2
GPT-5.5 100%

Source: Official The Aggregate Inversion and Execution Ladder leaderboard

GPT-5.2 on the official leaderboard passed 85% in December 2025

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% 2026 2027 Near the ceiling (85%) GPT-5.5 GPT-5.2 gpt-oss-120b 0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) GPT-5.5 GPT-5.2

Source: Official The Aggregate Inversion and Execution Ladder 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 Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (27 entries).
  • Pass Models scored 20 models we track, plus 7 other entries on its leaderboard.
  • Pass Newest score The newest is from November 2026, for Flint.
  • Note Frontier models 2 of the top 10 have a score, which is fine for a retired test. The strongest tested is GPT-6.2 Astra (#1).
  • Note Headroom Saturated. The best is 100%, past 85% since December 2025.
Leaderboard import 3 pass
  • Pass Importer We read the board on theaggregate.ai: 20 rows for models we track and 7 other entries 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