Orak Benchmark

What is Orak?

Orak (오락) is a foundational benchmark for evaluating Large Language Model (LLM) agents in diverse popular video games.

Released

2025-06-04

Evaluates

智能体, Agent, 任务执行, Task Execution, 不支持, Unsupported

Openness

unknown

Importer

OpenCompass Hub

Reported scores

0

Source provenance

• Original evidence https://hub.opencompass.org.cn/dataset-detail/Orak

Orak paper and code

• Paper https://arxiv.org/pdf/2506.03610
• Code https://github.com/krafton-ai/Orak

Which sources cite Orak?

• Orak

Related 智能体, Agent, 任务执行, Task Execution, 不支持, Unsupported benchmarks

• A-Bench
• A-OKVQA
• AbsPyramid
• ActionAtlas
• Ada-LEval
• AECBench
• AFQMC
• AgentBoard