AgentBench

Name: AgentBench
Rating: 0.9 (1800 reviews)
Author: THUDM

by THUDM1.8k

Benchmark for evaluating LLMs as agents

About

Benchmark for evaluating LLMs as agents Provides comprehensive functionality for seamless integration.

pip install agentbench

Community

Arxiv research paper access

Framework for developing applications powered by language models

Data framework for LLM applications with RAG capabilities