Benchmark Almanac

τ-bench 2.0

Not in model cards

Agents serve simulated customers in airline, retail, telecom and banking domains in six languages, by text or voice. Release 2.0 adds frozen task manifests and an intake domain for entity capture.

Sierra says 2.0 scores are not comparable with 1.x. The user simulator moved to GPT-5.5 and communication checks became LLM-judged. Languages are en, es, pt, hi, zh and ko.

Made by Sierra Released Oct 2026 Customer service All tasks public
Level
University Undergraduates and entry-level professionals
Scored
Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
Answers
Agent task Many steps in a terminal, browser, desktop, app, game or robot.
sierra-research/tau2-bench 2,366 stars Paper Leaderboard taubench.com

No scores here yet

We have no scores for τ-bench 2.0 yet.

See the live leaderboard on taubench.com

No lab model card reports it yet

We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.

Versions

We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.

Anthropic OpenAI Moonshot AI Qwen MiniMax Inception Labs
tau3-bench Passed 85% in Mar 2026, by Qwen3.5 397B-A17B
Mercury 2.5 96%
tau-bench Replaced by tau2-bench, Sep 2025
Claude Opus 4 70.5%
tau2-bench Passed 85% in Oct 2025, by MiniMax-M2
GPT-5.4 98.9%
tau3-bench Passed 85% in Mar 2026, by Qwen3.5 397B-A17B
Mercury 2.5 96%
Replaces tau3-bench

More in Tool use

All 125 in Tool use
  • AutomationBench In use Agent completes end-to-end business workflows inside a simulated company exposing dozens of REST APIs. Graded on resulting state. Gemini 4 Carbon 53.9%
  • MCP Atlas In use Multi-step workflows over real MCP servers. Scores tool-use competency on realistic tasks. Muse Spark 1.1 88.1%
  • BFCL V4 In use Models call functions and tools, including multi-turn agentic tasks with web search and memory. Checked against expected calls and outcomes. Qwen3.7-Max 75%
  • τ³-bench Banking In use Agents act as banking customer-service reps over many turns with a simulated user, following policy. Pass^1 on the final database state. Qwen 3.8 Max 55.1%
  • Claw-Eval In use 300 human-verified agent tasks (service orchestration, multimodal, multi-turn dialogue) audited via traces and snapshots. Rubric scores with Pass^3 consistency. Claude Sonnet 4.6 81.4%
  • DeepSearchQA In use Questions that need multi-step search and synthesis. Final answers graded against references. Kimi K3 95%