What it measures

Evaluates LLMs directly under live Codeforces Div. 1/2/3 contest conditions with human-calibrated Elo ratings.

Reading the measurement

A preference result depends on the opponents, judges and prompts in the comparison pool. Use it to study which outputs people or judges favor, then check correctness and reliability with a separate task-based evaluation. Ratings from different arenas do not share a universal scale.

Published evaluation

Overall Elo rating. Parenthesized human percentile ranks are not added to the rating.

Where this fits in an evaluation suite

The agent wrapper is part of the system. Pin tools, dependencies, timeouts, tests, and the repository revision. Compare contest coding separately from repository maintenance.

Start with a capability hypothesis, then choose a second test with a different failure mode. Keep the checkpoint, reasoning effort, agent tools, sampling policy, and scoring procedure fixed when measuring a training intervention. Record failures and missing outputs in the denominator.

A reproducible comparison

Overall Elo rating. Parenthesized human percentile ranks are not added to the rating.

Publish the exact run configuration, item coverage, score units, and available uncertainty. If a source supplies only an aggregate or a single run, preserve that limit. A new model name or a rounded score alone does not establish a capability improvement.

PRIMARY EVIDENCE & FURTHER READING

  1. CodeElo: Benchmarking Competition-level Code Generation of LLMs with Human-comparable Elo Ratings — published result table ↗

    Shanghaoran Quan, Jiaxi Yang, Bowen Yu, Bo Zheng, Dayiheng Liu, An Yang et al. · 2025

    All 17 authors

    Shanghaoran Quan, Jiaxi Yang, Bowen Yu, Bo Zheng, Dayiheng Liu, An Yang, Xuancheng Ren, Bofei Gao, Yibo Miao, Yunlong Feng, Zekun Wang, Jian Yang, Zeyu Cui, Yang Fan, Yichang Zhang, Binyuan Hui, Junyang Lin

Reviewed 2026-10-26. Explanation and interpretation by RLTasks. Benchmark ownership remains with its original authors.