New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
τ-bench benchmark
AI model leaderboard for the τ-bench benchmark. Compare how large language models score on τ-bench, see the full ranking, and understand what this AI benchmark measures. Multi-turn agent benchmark simulating realistic customer-service tool environments.
Leaderboard
No model scores reported yet.
Frequently asked questions about τ-bench
What is the τ-bench benchmark?
Multi-turn agent benchmark simulating realistic customer-service tool environments.
How is the τ-bench benchmark scored?
τ-bench is scored using the pass^1 (%) metric, where a higher score is better. The maximum achievable score is 100.000. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.
Which AI model scores highest on τ-bench?
No model scores have been reported for τ-bench yet. Check back as the leaderboard is updated.
Is a higher τ-bench score better?
Yes. On τ-bench a higher score indicates better performance, so models near the top of the leaderboard are the strongest.