Asteroid numbers
Name 48 freshly sampled asteroids' catalogue numbers, from memory.
Run Sep 9, 2026, 48 items.
Method
A follow-up to the exotic recall bank, where DeepSeek and Gemini beat the Claude models on asteroids. 48 named asteroids were drawn at random from NASA/JPL's catalogue (numbers 1,000 to 20,000, first observed before 2000), and each model gave the permanent number for each name, closed book, graded by exact match. The reversal held on the fresh sample.
A reply cut off by the output limit counts as a miss; that happened at most once per model. Reasoning effort was set to high.
Results
| # | Model | Accuracy |
|---|---|---|
| 1 | DeepSeek V4 Pro | 58.3% |
| 2 | Gemini 3.7 Flash | 56.2% |
| 3 | GPT-5.6 Sol | 52.1% |
| 4 | GPT-6 Astra | 47.9% |
| 5 | Claude Fable 5.1 | 33.3% |
| 6 | Claude Fable 5 | 27.1% |
| 7 | Claude Opus 5 | 20.8% |
In the main leaderboard
These results are collected for the main leaderboard alongside external benchmarks. New scores enter the rating through the regular audited data updates. Comparisons with that rating therefore share some of its inputs.
Interactive version: theaggregate.ai/our-benchmarks/asteroid-numbers · How It Works · Data refreshed daily, snapshot 2026-10-11.