OIBench: 250 Olympiad-Level Informatics Problems

OIBench is a high-quality, private, and challenging benchmark consisting of 250 carefully curated original problems. The dataset contains algorithm problem statements, solutions, and associated metadata such as test cases, pseudo code, and difficulty levels. It was created by meituan-longcat and last updated on July 15, 2025.

Use Cases

• Training code generation models based on the 250 original problem statements.
• Benchmarking AI performance on challenging, olympiad-level informatics tasks.
• Analyzing solution strategies and algorithmic patterns using the provided pseudo code and solutions.
• Evaluating the difficulty and complexity of algorithm problems using the associated metadata.

Strengths

• Contains 250 carefully curated original problems.
• Includes multiple associated data points per problem, such as test cases, pseudo code, and difficulty levels.
• Stored in Parquet format for efficient access and analysis.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Freshness should be verified as the last update was July 15, 2025.

Provenance

Source

huggingface

Collection Method

Curated and processed by meituan-longcat.

Freshness

Last updated 2025-07-15 09:28:19.