ExplorationBench - AlienCode (Open-Book): leaderboard
Metric: Held-out accuracy with the complete rule set supplied before exploration (%; open-book control, one run per system, no tool calls; AlienCode, 70 program-synthesis and prediction tasks graded by an interpreter on private inputs; 70 held-out tasks each answered three times in a tool-disabled copy of the conversation and averaged; highest reasoning setting each API offers). Source: arxiv.org. Saturation forecast: Around December 2026. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5 (Max) | 84.3 |
| 2 | Gemini 3.8 Flash (High) | 78.1 |
| 3 | GPT-5.6 Sol (Max) | 68.1 |
| 4 | Kimi K3 (Max) | 52.4 |
| 5 | Grok 4.6 (xHigh) | 51 |
| 6 | DeepSeek V4.1 Flash (Max) | 39.5 |
| 7 | DeepSeek V4 Pro (Max) | 37.6 |
Interactive version: theaggregate.ai/benchmark?slug=explorationbench-aliencode-open-book · How It Works · Data refreshed daily, snapshot 2026-09-26.