AgentBench: Evaluate LLMs as Agents

Active

A benchmark designed to evaluate LLMs as Agents

Publisher: Tsinghua University
Domain: Coding
License: mit
Published: Aug 2025
Notable for: Benchmark for evaluating Coding.
Canonical: github.com/UKGovernmentBEIS/inspect_evals/tree/main/src/inspect_evals/agent_bench

Cite

Notes

Only stored in your browser.

Attribution

README: github.com/UKGovernmentBEIS/inspect_evals/blob/main/src/inspect_evals/agent_bench/eval.yamlMIT

Attribution policy →

FAQ

What is AgentBench: Evaluate LLMs as Agents?: A benchmark designed to evaluate LLMs as Agents
What license is AgentBench: Evaluate LLMs as Agents under?: AgentBench: Evaluate LLMs as Agents is available under mit.