reasoning benchmark
Tau2 Telecom
τ²-Bench telecom domain evaluates conversational agents in a dual-control environment modeled as a Dec-POMDP, where both agent and user use tools in shared telecommunications troubleshooting scenarios that test coordination and communication capabilities.
Updated Aug 7, 2026
Published models35
Registry coverage35
MetricScore
EvidenceB
Tau2 Telecom leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
Score distribution
Top published rows on the benchmark's original scale.
Tau2 Telecom highlights
The top published results on this benchmark's own scale.
What is Tau2 Telecom?
Definition and scoring fields from the benchmark registry.
τ²-Bench telecom domain evaluates conversational agents in a dual-control environment modeled as a Dec-POMDP, where both agent and user use tools in shared telecommunications troubleshooting scenarios that test coordination and communication capabilities.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- Tau2 Telecom
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- tau2-telecom|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about Tau2 Telecom.
Which model scores highest on Tau2 Telecom?
Claude Opus 4.6 is currently ranked first with 99.3%.
What does Tau2 Telecom measure?
τ²-Bench telecom domain evaluates conversational agents in a dual-control environment modeled as a Dec-POMDP, where both agent and user use tools in shared telecommunications troubleshooting scenarios that test coordination and communication capabilities.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
35 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.