Factorio Learning Environment - Automation Milestones: leaderboard
Factorio Learning Environment (FLE) automation milestones: the count the FLE leaderboard lists beside overall milestones for LLM agents that write Python programs to build Factorio factories; the leaderboard defines it only as automation milestones reached. No fixed maximum; higher is better.
Metric: Automation milestones reached. Source: jackhopkins.github.io. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude 3.5 Sonnet | 13 | #362 |
| 2 | GPT-4o | 9 | #357 |
| 3 | DeepSeek V3 | 7 | #329 |
| 4 | Gemini 2.0 Flash | 6 | #351 |
| 5 | GPT-4o Mini | 4 | #598 |
| 6 | Llama 3.3 70B | 4 | #578 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=factorio-learning-environment-automation-milestones · How It Works · Data refreshed daily, snapshot 2026-11-05.