reasoning benchmark
Tau2 Retail Leaderboard
τ²-bench retail domain evaluates conversational AI agents in customer service scenarios within a dual-control environment where both agent and user can interact with tools. Tests tool-agent-user interaction, rule adherence, and task consistency in retail customer support contexts.
Updated Aug 17, 2026
Models26
Model coverage26
MetricScore
EvidenceB
Tau2 Retail Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAN | Score91.9% | Percentile100.0% | Participants26 | EvidenceC | Evaluated |
| Rank02 | ModelAN | Score91.7% | Percentile96.0% | Participants26 | EvidenceC | Evaluated |
| Rank03 | ModelAN | Score88.9% | Percentile92.0% | Participants26 | EvidenceC | Evaluated |
| Rank04 | ModelME | Score88.6% | Percentile88.0% | Participants26 | EvidenceC | Evaluated |
| Rank05 | ModelAN | Score83.2% | Percentile84.0% | Participants26 | EvidenceC | Evaluated |
| Rank06 | ModelOP | Score82.0% | Percentile80.0% | Participants26 | EvidenceC | Evaluated |
| Rank07 | ModelOP | Score81.1% | Percentile76.0% | Participants26 | EvidenceC | Evaluated |
| Rank08 | ModelOP | Score80.2% | Percentile72.0% | Participants26 | EvidenceC | Evaluated |
| Rank09 | ModelAM Nova 2 OmniAmazon | Score78.3% | Percentile68.0% | Participants26 | EvidenceC | Evaluated |
| Rank10 | ModelOP | Score77.9% | Percentile64.0% | Participants26 | EvidenceC | Evaluated |
| Rank11 | ModelOP | Score77.9% | Percentile60.0% | Participants26 | EvidenceC | Evaluated |
| Rank12 | ModelOP | Score77.9% | Percentile56.0% | Participants26 | EvidenceC | Evaluated |
| Rank13 | ModelAMNova 2 ProAmazon | Score77.7% | Percentile52.0% | Participants26 | EvidenceC | Evaluated |
| Rank14 | ModelAMNova 2 LiteAmazon | Score76.5% | Percentile48.0% | Participants26 | EvidenceC | Evaluated |
| Rank15 | ModelME | Score73.1% | Percentile44.0% | Participants26 | EvidenceC | Evaluated |
| Rank16 | ModelAC | Score71.9% | Percentile40.0% | Participants26 | EvidenceC | Evaluated |
| Rank17 | ModelME | Score71.5% | Percentile36.0% | Participants26 | EvidenceC | Evaluated |
| Rank18 | ModelAC | Score71.3% | Percentile32.0% | Participants26 | EvidenceC | Evaluated |
| Rank19 | ModelME | Score71.3% | Percentile28.0% | Participants26 | EvidenceC | Evaluated |
| Rank20 | ModelMA | Score70.6% | Percentile24.0% | Participants26 | EvidenceC | Evaluated |
| Rank21 | ModelMA | Score70.6% | Percentile20.0% | Participants26 | EvidenceC | Evaluated |
| Rank22 | ModelAC | Score67.8% | Percentile16.0% | Participants26 | EvidenceC | Evaluated |
| Rank23 | ModelOP | Score63.4% | Percentile12.0% | Participants26 | EvidenceC | Evaluated |
| Rank24 | ModelNV | Score62.8% | Percentile8.0% | Participants26 | EvidenceC | Evaluated |
| Rank25 | ModelAC | Score57.3% | Percentile4.0% | Participants26 | EvidenceC | Evaluated |
| Rank26 | ModelNV | Score56.9% | Percentile0.0% | Participants26 | EvidenceC | Evaluated |
Tau2 Retail Highlights
The leading models and scores on this benchmark.
Tau2 Retail Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for Tau2 Retail
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis tau2 retail AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01ANClaude Opus 4.6Anthropic
- Score
- 91.9%
- Price
- $5.0 input / $25 output per 1M tokens
- Speed
- Up to 17 tok/s via Anthropic
Strengths
- Ranks #1 of 26 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures Tau2 Retail, not total model capability
- 02ANClaude Sonnet 4.6Anthropic
- Score
- 91.7%
- Price
- $3.0 input / $15 output per 1M tokens
- Speed
- Up to 15 tok/s via Anthropic
Strengths
- Ranks #2 of 26 compared models
- 96th percentile on this benchmark
- C evidence result
Considerations
- This result measures Tau2 Retail, not total model capability
- 03ANClaude Opus 4.5Anthropic
- Score
- 88.9%
- Price
- $5.0 input / $25 output per 1M tokens
- Speed
- Up to 42 tok/s via Anthropic
Strengths
- Ranks #3 of 26 compared models
- 92th percentile on this benchmark
- C evidence result
Considerations
- This result measures Tau2 Retail, not total model capability
- 04MELongCat-Flash-Thinking-2601Meituan
- Score
- 88.6%
- Speed
- Up to 100 tok/s via Meituan
Strengths
- Ranks #4 of 26 compared models
- 88th percentile on this benchmark
- C evidence result
Considerations
- This result measures Tau2 Retail, not total model capability
- 05ANClaude Haiku 4.5Anthropic
- Score
- 83.2%
- Price
- $1.0 input / $5.0 output per 1M tokens
- Speed
- Up to 100 tok/s via Anthropic
Strengths
- Ranks #5 of 26 compared models
- 84th percentile on this benchmark
- C evidence result
Considerations
- This result measures Tau2 Retail, not total model capability
Selection summary
Best AI Models for Tau2 Retail
Claude Opus 4.6 currently leads Tau2 Retail with 91.9%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is Tau2 Retail?
What Tau2 Retail measures and how its scores work.
τ²-bench retail domain evaluates conversational AI agents in customer service scenarios within a dual-control environment where both agent and user can interact with tools. Tests tool-agent-user interaction, rule adherence, and task consistency in retail customer support contexts.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- Tau2 Retail
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- tau2-retail|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about Tau2 Retail.
Which model scores highest on Tau2 Retail?
Claude Opus 4.6 is currently ranked first with 91.9%.
What does Tau2 Retail measure?
τ²-bench retail domain evaluates conversational AI agents in customer service scenarios within a dual-control environment where both agent and user can interact with tools. Tests tool-agent-user interaction, rule adherence, and task consistency in retail customer support contexts.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
26 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.