# What are the main types of agent benchmarks?

Six, split by what the agent is asked to do: pick the right tool call, hold a conversation against a policy, fix a real bug, operate inside a changing environment, build something as a team, or score how well a judge model grades others.

Tool-calling benchmarks like [BFCL](/answers/bfcl/what-is-bfcl) check whether a model picks the right function and fills in the right arguments. Conversational benchmarks like [tau-bench](/answers/tau-bench/what-is-tau-bench) run a simulated user against a policy document and score only the account changes that resulted. Coding benchmarks like [SWE-bench Pro](/answers/swe-bench/what-is-swe-bench-pro) hand the agent a real bug report and grade it against the project's own tests, while environment benchmarks like [Terminal-Bench](/answers/terminal-bench/what-is-terminal-bench) and [Gaia2](/answers/gaia/what-is-gaia2) drop it into a shell to finish a job, or a simulated phone where time keeps passing, and check what state it leaves behind. The last two score different units: one grades a whole team of agents building something end to end, and the other benchmarks the judges, where a judge's own rank can [shift by up to 14 positions](/answers/judge-bias-research/do-llm-judge-rankings-transfer-across-benchmarks) from one benchmark to the next.

A tool-calling score doesn't predict how an agent holds a conversation, and neither predicts how it behaves on your own traffic.

See also: [What is OSWorld?](/answers/osworld/what-is-osworld)

---

Source: https://tessary.ai/answers/agent-benchmarks/what-are-the-main-types-of-agent-benchmarks
More on Agent benchmarks: https://tessary.ai/answers/agent-benchmarks
From Tessary, agent reliability for AI agents in production: https://tessary.ai