Ayanami0730/deep_research_bench
A benchmark suite for evaluating AI agents on deep research tasks with automatic scoring and a public leaderboard.
Not currently ranked — collecting fresh signals.
star history
DeepResearch Bench is a comprehensive evaluation framework for assessing AI agents capable of deep research. The project provides a dataset of research tasks with human-annotated reference answers, an automatic evaluator using frontier models (GPT-5.5, GPT-5.4-mini, Gemini-2.5-Pro) to score agent outputs on dimensions like Overall quality, PAR, and FAS, and a public leaderboard hosted on Hugging Face for comparing agent performance.
Frequently asked
- What is Ayanami0730/deep_research_bench?
- A benchmark suite for evaluating AI agents on deep research tasks with automatic scoring and a public leaderboard.
- Is deep_research_bench open source?
- Yes — Ayanami0730/deep_research_bench is open source, released under the Apache-2.0 license.
- What language is deep_research_bench written in?
- Ayanami0730/deep_research_bench is primarily written in Python.
- How popular is deep_research_bench?
- Ayanami0730/deep_research_bench has 795 stars on GitHub.
- Where can I find deep_research_bench?
- Ayanami0730/deep_research_bench is on GitHub at https://github.com/Ayanami0730/deep_research_bench.