PECC - Advent of Code: leaderboard
The Advent of Code track of PECC (arXiv) — Problem Extraction & Coding Challenges from Advent of Code that require reading comprehension before coding; solve rate (%).
Metric: Solve rate (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-3.5 Turbo | 50 | #865 |
| 2 | Mixtral 8x7B Instruct (v0.1) | 15.31 | #1073 |
| 3 | Phi-3 Mini 128K Instruct | 10.13 | #1298 |
| 4 | WizardLM-2-7B | 5.87 | #1395 |
| 5 | Mistral Instruct v0.1 7B | 3 | #1454 |
| 6 | Llama 3 8B Instruct | 1.53 | #1129 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=pecc-advent-of-code · How It Works · Data refreshed daily, snapshot 2026-10-20.