AstroReason-Bench - Revisit Optimization - Revisit Gap: leaderboard
Metric: Mean revisit gap over monitoring targets (hours; one run of five procedurally generated cases, Claude Code ReAct harness). Source: arxiv.org. Saturation forecast: Around February 2027. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 | 18.83 |
| 2 | Gemini 3 Flash | 24.96 |
| 3 | DeepSeek V3.1 NEX N1 | 26.78 |
| 4 | DeepSeek V3.2 | 29.89 |
| 5 | Qwen3 Coder | 38.58 |
Interactive version: theaggregate.ai/benchmark?slug=astroreason-bench-revisit-optimization-revisit-gap · How It Works · Data refreshed daily, snapshot 2026-09-25.