Sign in
← Back to search

sierra-research/tau2-bench

τ-Bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains

#ai#benchmark#conversational-agents#language-model-agent#llm
Sign in to like sierra-research/tau2-bench Open website Open GitHub
Stars
1,230
Forks
318
Commits
147
Language
Python
Awesome lists
1

Similar repositories

THUDM/AgentBench

A Comprehensive Benchmark to Evaluate LLMs as Agents (ICLR'24)

3451 stars
Python 4 awesome lists

Tracked growth

1 capture since 2026-05-25

Latest capture 2026-05-25 20:57

Stars history

Total stars

Commits history

Default branch commits

Metadata

  • Created: 2025-06-09
  • First commit: —
  • Last pushed: 2026-05-21
  • Website: https://www.taubench.com
  • Archived: no
  • Stack detected: —
  • License: MIT

AI development signals

AI agent config detected

12 config paths 10 files 2 directories
Agent instructions 7 Cursor 5

Key config paths

  • dir .cursor
  • file AGENTS.md
  • file src/tau2/agent/AGENTS.md
  • file src/tau2/domains/AGENTS.md
  • file src/tau2/evaluator/AGENTS.md
  • file src/tau2/voice/AGENTS.md

2 more config paths detected.

Review config paths
  • Cursor .cursor
  • Cursor .cursor/rules
  • Cursor .cursor/rules/audio-native-provider.md
  • Cursor .cursor/rules/background-audio-files.md
  • Cursor .cursor/rules/nova-sonic.md
  • Agent instructions AGENTS.md
  • Agent instructions src/tau2/agent/AGENTS.md
  • Agent instructions src/tau2/domains/AGENTS.md
  • Agent instructions src/tau2/evaluator/AGENTS.md
  • Agent instructions src/tau2/voice/AGENTS.md
  • Agent instructions src/tau2/voice/audio_native/AGENTS.md
  • Agent instructions tests/AGENTS.md