reasoning benchmark
Tau2 Telecom Leaderboard
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 17, 2026
Models35
Model coverage35
MetricScore
EvidenceB
Tau2 Telecom Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAN | Score99.3% | Percentile100.0% | Participants35 | EvidenceC | Evaluated |
| Rank02 | ModelME | Score99.3% | Percentile97.1% | Participants35 | EvidenceC | Evaluated |
| Rank03 | ModelOP | Score98.9% | Percentile94.1% | Participants35 | EvidenceC | Evaluated |
| Rank04 | ModelOP | Score98.7% | Percentile91.2% | Participants35 | EvidenceC | Evaluated |
| Rank05 | ModelAN | Score98.2% | Percentile88.2% | Participants35 | EvidenceC | Evaluated |
| Rank06 | ModelOP | Score98.0% | Percentile85.3% | Participants35 | EvidenceC | Evaluated |
| Rank07 | ModelAN | Score97.9% | Percentile82.3% | Participants35 | EvidenceC | Evaluated |
| Rank08 | ModelXI | Score96.8% | Percentile79.4% | Participants35 | EvidenceC | Evaluated |
| Rank09 | ModelOP | Score96.7% | Percentile76.5% | Participants35 | EvidenceC | Evaluated |
| Rank10 | ModelOP | Score95.6% | Percentile73.5% | Participants35 | EvidenceC | Evaluated |
| Rank11 | ModelOP | Score95.6% | Percentile70.6% | Participants35 | EvidenceC | Evaluated |
| Rank12 | ModelOP | Score95.6% | Percentile67.7% | Participants35 | EvidenceC | Evaluated |
| Rank13 | ModelOP | Score93.4% | Percentile64.7% | Participants35 | EvidenceC | Evaluated |
| Rank14 | ModelAM Nova 2 ProAmazon | Score92.7% | Percentile61.8% | Participants35 | EvidenceC | Evaluated |
| Rank15 | ModelOP | Score92.5% | Percentile58.8% | Participants35 | EvidenceC | Evaluated |
| Rank16 | ModelME | Score91.5% | Percentile55.9% | Participants35 | EvidenceC | Evaluated |
| Rank17 | ModelMI | Score87.0% | Percentile52.9% | Participants35 | EvidenceC | Evaluated |
| Rank18 | ModelMI | Score87.0% | Percentile50.0% | Participants35 | EvidenceC | Evaluated |
| Rank19 | ModelCO | Score85.0% | Percentile47.1% | Participants35 | EvidenceC | Evaluated |
| Rank20 | ModelME | Score83.1% | Percentile44.1% | Participants35 | EvidenceC | Evaluated |
| Rank21 | ModelAN | Score83.0% | Percentile41.2% | Participants35 | EvidenceC | Evaluated |
| Rank22 | ModelAMNova 2 OmniAmazon | Score80.0% | Percentile38.2% | Participants35 | EvidenceC | Evaluated |
| Rank23 | ModelAMNova 2 LiteAmazon | Score76.0% | Percentile35.3% | Participants35 | EvidenceC | Evaluated |
| Rank24 | ModelME | Score73.7% | Percentile32.4% | Participants35 | EvidenceC | Evaluated |
| Rank25 | ModelME | Score72.8% | Percentile29.4% | Participants35 | EvidenceC | Evaluated |
| Rank26 | ModelMI | Score71.7% | Percentile26.5% | Participants35 | EvidenceC | Evaluated |
| Rank27 | ModelMA | Score65.8% | Percentile23.5% | Participants35 | EvidenceC | Evaluated |
| Rank28 | ModelMA | Score65.8% | Percentile20.6% | Participants35 | EvidenceC | Evaluated |
| Rank29 | ModelNV | Score64.4% | Percentile17.6% | Participants35 | EvidenceC | Evaluated |
| Rank30 | ModelOP | Score58.2% | Percentile14.7% | Participants35 | EvidenceC | Evaluated |
| Rank31 | ModelAC | Score45.6% | Percentile11.8% | Participants35 | EvidenceC | Evaluated |
| Rank32 | ModelAC | Score43.9% | Percentile8.8% | Participants35 | EvidenceC | Evaluated |
| Rank33 | ModelNV | Score42.2% | Percentile5.9% | Participants35 | EvidenceC | Evaluated |
| Rank34 | ModelOP | Score23.5% | Percentile2.9% | Participants35 | EvidenceC | Evaluated |
| Rank35 | ModelAC | Score13.2% | Percentile0.0% | Participants35 | EvidenceC | Evaluated |
Tau2 Telecom Highlights
The leading models and scores on this benchmark.
Tau2 Telecom Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for Tau2 Telecom
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis tau2 telecom 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
- 99.3%
- Price
- $5.0 input / $25 output per 1M tokens
- Speed
- Up to 17 tok/s via Anthropic
Strengths
- Ranks #1 of 35 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures Tau2 Telecom, not total model capability
- 02MELongCat-Flash-Thinking-2601Meituan
- Score
- 99.3%
- Speed
- Up to 100 tok/s via Meituan
Strengths
- Ranks #2 of 35 compared models
- 97th percentile on this benchmark
- C evidence result
Considerations
- This result measures Tau2 Telecom, not total model capability
- 03OPGPT-5.4OpenAI
- Score
- 98.9%
- Price
- $2.5 input / $15 output per 1M tokens
- Speed
- Up to 9.1 tok/s via OpenAI
Strengths
- Ranks #3 of 35 compared models
- 94th percentile on this benchmark
- C evidence result
Considerations
- This result measures Tau2 Telecom, not total model capability
- 04OPGPT-5.2OpenAI
- Score
- 98.7%
- Price
- $1.8 input / $14 output per 1M tokens
- Speed
- Up to 100 tok/s via OpenAI
Strengths
- Ranks #4 of 35 compared models
- 91th percentile on this benchmark
- C evidence result
Considerations
- This result measures Tau2 Telecom, not total model capability
- 05ANClaude Opus 4.5Anthropic
- Score
- 98.2%
- Price
- $5.0 input / $25 output per 1M tokens
- Speed
- Up to 42 tok/s via Anthropic
Strengths
- Ranks #5 of 35 compared models
- 88th percentile on this benchmark
- C evidence result
Considerations
- This result measures Tau2 Telecom, not total model capability
Selection summary
Best AI Models for Tau2 Telecom
Claude Opus 4.6 currently leads Tau2 Telecom with 99.3%. 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 Telecom?
What Tau2 Telecom measures and how its scores work.
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. This benchmark is not independently verified and has an evidence level of B.
- Family
- Tau2 Telecom
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- tau2-telecom|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
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 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.