PowerBench

What is PowerBench?

PowerBench evaluates LLM agents on 300 questions across three task families requiring autonomous evidence retrieval and reasoning over interconnected power…

Released

2026-09-28

Evaluates

General AI, Information retrieval, Language & Knowledge, Reasoning

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2609.34492

PowerBench paper and code

• Paper https://arxiv.org/abs/2609.34492
• Code https://github.com/open-compass/PowerBench

Which sources cite PowerBench?

• PowerBench

Related General AI, Information retrieval, Language & Knowledge, Reasoning benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 2d-fet-bench
• 3DCodeBench
• 5-Dialects-BN
• A Large-Scale Dataset and Benchmark
• A Real-World CAD Segmentation Benchmark
• A-OKVQA
• A3S-Bench