tau-bench

tau-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains

↓ tau2-bench

Simulated customer-service conversations in retail (115 tasks) and airline (50 tasks) domains where the agent must talk to an LLM-simulated user, call database APIs and follow a written policy. A task passes only if the final database state matches the annotated goal; pass^k measures whether the same task succeeds in all k independent trials, exposing agent inconsistency. The reference benchmark for policy-following tool agents.

69.2%
TC (claude-3-5-sonnet-20241022), retail
official
2024-06-17 → 2024-10-22: 35.2% → 69.2%
Released
2024-06
Maintainer
Sierra Research (Yao et al.)
Status
retired
Contamination
high
Metric
pass^1 (percent, ↑)
Tasks
165
Domains
tool-use instruction-following general-assistant
human
no measured baseline

Notes. The repository README now warns that its airline and retail tasks are outdated and directs users to the tau2-bench repository (tau^3-bench) for fixed tasks; developer reports of 'tau-bench' after mid-2025 usually mean tau2-bench. Scores depend on the user-simulator model, which is not standardised.

Full ledger

SystemDeveloperScoreDateSourceConditions
TC (claude-3-5-sonnet-20241022), retailAnthropic69.2%officialsplit: retail scaffold: function calling (TC)
Repository README leaderboard; date is the model's release date (README gives none).
gpt-4o (function calling), tau-retailOpenAI61.2%papersplit: retail scaffold: function calling (TC)
Table 2 of the paper; user simulated by gpt-4.
TC (claude-3-5-sonnet-20241022), airlineAnthropic46%officialsplit: airline scaffold: function calling (TC)
Repository README leaderboard; date is the model's release date (README gives none).
gpt-4o (function calling), tau-airlineOpenAI35.2%papersplit: airline scaffold: function calling (TC)
Table 2 of the paper; user simulated by gpt-4.