AgentBench

🌐 Website | 🐦 Twitter | ✉️ Google Group | 📃 Paper

👋 Join our Slack for Q & A or collaboration on next version of AgentBench!

📌Introducing AgentBench v0.2🎉

You are now browsing AgentBench v0.2. If you wish to use the older version, you can revert to v0.1.

Based on v0.1, we:

• Updated the framework architecture for easier use and extension
• Adjusted some task settings
• Added test results for more models
• Released the full data for the Dev and Test sets

AgentBench: Evaluating LLMs as Agents

https://github.com/THUDM/AgentBench/assets/129033897/656eed6e-d9d9-4d07-b568-f43f5a451f04

AgentBench is the first benchmark designed to evaluate LLM-as-Agent across a diverse spectrum of different environments. It encompasses 8 distinct environments to provide a more comprehensive evaluation of the LLMs' ability to operate as autonomous agents in various scenarios. These environments include 5 freshly created domains, namely

• Operating System (OS)
• Database (DB)
• Knowledge Graph (KG)
• Digital Card Game (DCG)
• Lateral Thinking Puzzles (LTP)

as well as 3 recompiled from published datasets:

• House-Holding (HH) (ALFWorld)
• Web Shopping (WS) (WebShop)
• Web Browsing (WB) (Mind2Web)

Table of Contents

• Dataset Summary
• Leaderboard
• Quick Start
• Next Steps
• Citation

Dataset Summary

We offer two splits for each dataset: Dev and Test. The multi-turn interaction requires an LLMs to generate around 4k and 13k times respectively.

Leaderboard

Here is the scores on test set (standard) results of AgentBench.

While LLMs begin to manifest their proficiency in LLM-as-Agent, gaps between models and the distance towards practical usability are significant.

Quick Start

This section will guide you on how to quickly use gpt-3.5-turbo-0613 as an agent to launch the dbbench-std and os-std tasks. For the specific framework structure, please refer to Framework Introduction. For more detailed configuration and launch methods, please check Configuration Guide and Program Entrance Guide.

Step 1. Prerequisites

Clone this repo and install the dependencies.

cd AgentBench
conda create -n agent-bench python=3.9
conda activate agent-bench
pip install -r requirements.txt

Ensure that Docker is properly installed.

docker ps

Build required images for dbbench-std and os-std.

docker pull mysql
docker pull ubuntu
docker build -f data/os_interaction/res/dockerfiles/default data/os_interaction/res/dockerfiles --tag local-os/default
docker build -f data/os_interaction/res/dockerfiles/packages data/os_interaction/res/dockerfiles --tag local-os/packages
docker build -f data/os_interaction/res/dockerfiles/ubuntu data/os_interaction/res/dockerfiles --tag local-os/ubuntu

Step 2. Configure the Agent

Fill in your OpenAI API Key at the correct location in configs/agents/openai-chat.yaml. (e.g. gpt-3.5-turbo-0613)

You can try using python -m src.client.agent_test to check if your agent is configured correctly.

By default, gpt-3.5-turbo-0613 will be started. You can replace it with other agents by modifying the parameters:

python -m src.client.agent_test --config configs/agents/api_agents.yaml --agent gpt-3.5-turbo-0613

Step 3. Start the task server

Starting the task worker involves specific tasks. Manual starting might be cumbersome; hence, we provide an automated script.

The assumption for this step is that ports from 5000 to 5015 are available.

python -m src.start_task -a

This will launch five task_workers each for dbbench-std and os-std tasks and automatically connect them to the controller on port 5000. After executing this command, please allow approximately 1 minute for the task setup to complete.

Step 4. Start the assigner

This step is to actually start the tasks.

If everything is correctly configured so far, you can now initiate the task tests.

python -m src.assigner

Next Steps

If you wish to launch more tasks or use other models, you can refer to the content in Configuration Guide and Program Entrance Guide.

For the environment of the remaining five tasks, you will need to download the Docker images we provide.

longinyu/agentbench-ltp
longinyu/agentbench-webshop
longinyu/agentbench-mind2web
longinyu/agentbench-card_game
longinyu/agentbench-alfworld

The resource consumption of a single task_worker for the eight tasks is roughly as follows; consider this when launching:

Citation

@article{liu2023agentbench,
  title   = {AgentBench: Evaluating LLMs as Agents},
  author  = {Xiao Liu and Hao Yu and Hanchen Zhang and Yifan Xu and Xuanyu Lei and Hanyu Lai and Yu Gu and Hangliang Ding and Kaiwen Men and Kejuan Yang and Shudan Zhang and Xiang Deng and Aohan Zeng and Zhengxiao Du and Chenhui Zhang and Sheng Shen and Tianjun Zhang and Yu Su and Huan Sun and Minlie Huang and Yuxiao Dong and Jie Tang},
  year    = {2023},
  journal = {arXiv preprint arXiv: 2308.03688}
}