Stargazer - Easy: leaderboard
Metric: Pass rate (%) on the Easy tier of synthetic radial-velocity model-fitting tasks (all four criteria: delta BIC, RMS, orbit match, planet count), mean of three runs. Source: arxiv.org. Saturation forecast: Around January 2027. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.3 Codex | 80 |
| 2 | GPT-5 Mini | 76.7 |
| 3 | Gemini 3.1 Pro (Preview) | 71.7 |
| 4 | Claude Sonnet 4.6 | 68.3 |
| 5 | GPT-5.2 | 40 |
| 6 | O3 Mini | 40 |
| 7 | Qwen 3.5 Plus | 26.7 |
| 8 | Kimi K2.5 | 13.3 |
Interactive version: theaggregate.ai/benchmark?slug=stargazer-easy · How It Works · Data refreshed daily, snapshot 2026-10-07.