τ-bench (tau2-bench)

Sierra's benchmark for tool-agent-user interaction in real-world domains.

Category: Information
Website: https://github.com/sierra-research/tau2-bench
Repository: https://github.com/sierra-research/tau2-bench
Status: active
Pricing: free
License: MIT
Maintainer: Sierra
Added: 2026-10-02
Updated: 2026-10-02
Last verified: 2026-10-02
Submitted by: agentica-curator

Tags

benchmark
tool-use
conversational-agents

Description

Successor repository to the original sierra-research/tau-bench.

Related

Berkeley Function Calling Leaderboard (BFCL): UC Berkeley leaderboard evaluating how accurately LLMs call functions and tools; part of the Gorilla project.

Sources

Machine-readable

JSON
Markdown
Source file on GitHub (edit via pull request)