𝜏²-Bench Telecom Benchmark Leaderboard | Artificial Analysis
K
𝜏²-Bench Telecom Benchmark Leaderboard
A dual-control conversational AI benchmark simulating technical support scenarios where both agent and user must coordinate actions to resolve telecom service issues.
See example tasks
Background
Methodology
Publication
𝜏²-Bench: Evaluating Conversational Agents in a Dual-Control Environment
Victor Barres, Honghua Dong, Soham Ray, Xujie Si, Karthik Narasimhan.
Abstract
Existing benchmarks for conversational AI agents simulate single-control environments, where only the AI agent can use tools to interact with the world, while the user remains a passive information provider. This differs from real-world scenarios like technical support, where users need to actively participate in modifying the state of the (shared) world. In order to address this gap, we introduce 𝜏²-bench, with four key contributions: 1) A novel Telecom dual-control domain modeled as a Dec-POMDP, where both agent and user make use of tools to act in a shared, dynamic environment that tests both agent coordination and communication, 2) A compositional task generator that programmatically creates diverse, verifiable tasks from atomic components, ensuring domain coverage and controlled complexity, 3) A reliable user simulator tightly coupled with the environment, whose behavior is constrained by tools and observable states, improving simulation fidelity, 4) Fine-grained analysis of agent performance through multiple ablations including separating errors arising from reasoning vs communication/coordination. In particular, our experiments show significant performance drops when agents shift from no-user to dual-control, highlighting the challenges of guiding users. Overall, 𝜏²-bench provides a controlled testbed for agents that must both reason effectively and guide user actions.
Related Links
𝜏²-Bench Telecom
JT-35B-Flash scores the highest on 𝜏²-Bench Telecom with a score of 99.1%, followed by GLM-4.7-Flash (Reasoning) with a score of 98.8%, and Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) with a score of 98.5%
ScoreToken UsageScore vs. Release DateExample Tasks
Score
𝜏²-Bench Telecom Benchmark Leaderboard: Score
Independently benchmarked by Artificial Analysis
31 of 426 models
Add model from specific provider
Reasoning models are indicated by a lightbulb icon
Token Usage
𝜏²-Bench Telecom Benchmark Leaderboard: Output Token Usage
Output tokens used to run the evaluation
31 of 426 models
Add model from specific provider
Answer tokensReasoning tokens
Reasoning models are indicated by a lightbulb icon
Evaluation Output Tokens
Score vs. Release Date
𝜏²-Bench Telecom Benchmark Leaderboard: Score vs. Release Date
31 of 426 models
Add model from specific provider
Most attractive region
OpenAIMetaGoogleAnthropicMistralDeepSeekxAIAmazonUpstageMiniMaxNVIDIAKimiStepFunXiaomiMBZUAI Institute of Foundation ModelsChina MobileZ AIAlibaba
Example Tasks
Mobile Data Not Working
Scenario: Mobile data is not working properly. Either stops working or is very slow.
User Context: • Name: John Smith • Phone: 555-123-2002 • Location: At home in the United States • Constraints: Not willing to change mobile data plan but will refuel 2.0 GB if necessary. No wifi access.
Initial State: Mobile data is turned off and data limit exceeded (15.1 GB used)
Required Actions: • User turns mobile data on • Agent refuels 2.0 GB of data to customer account
Success Criteria (all must be met): ✓ Mobile data is turned on ✓ Internet speed test shows excellent (200+ Mbps) ✓ Exactly 2.0 GB of data refueled to account
Click to expand
No Service Connection Issue
Scenario: Phone showing "No Service" for the past few hours
User Context: • Name: John Smith • Phone: 555-123-2002 • Location: At home • Constraints: Retired librarian, 64 years old, limited tech knowledge, anxious about technology
Initial State: Airplane mode is on, SIM card is locked with PIN, and SIM card is physically unseated
Required Actions: • Agent attempts troubleshooting • Agent transfers to human agents when unable to resolve
Success Criteria (all must be met): ✓ Service status remains "No Service" ✓ Agent transfers to human support with summary "I cannot fix the issue."
Click to expand
MMS Issues While Abroad
Scenario: Unable to send MMS messages for the past few hours
User Context: • Name: John Smith • Phone: 555-123-2002 • Location: Abroad in France • Constraints: Willing to refuel 2.0 GB data but not change plan
Initial State: Airplane mode is on, messaging app lacks SMS and storage permissions, data limit exceeded (15.1 GB used), and roaming is disabled on both device and account
Required Actions: • User turns airplane mode off • User grants SMS permission to messaging app • User grants storage permission to messaging app • Agent refuels 2.0 GB of data • Agent enables roaming on customer account • User turns roaming on in device settings
Success Criteria (all must be met): ✓ MMS messages can be sent successfully ✓ Exactly 2.0 GB of data refueled to account
Click to expand
Explore Evaluations
Artificial Analysis Intelligence IndexA composite benchmark aggregating ten challenging evaluations to provide a holistic measure of AI capabilities across mathematics, science, coding, and reasoning.GDPval-AA LeaderboardGDPval-AA is Artificial Analysis' evaluation framework for OpenAI's GDPval dataset. It tests AI models on real-world tasks across 44 occupations and 9 major industries. Models are given shell access and web browsing capabilities in an agentic loop via Stirrup to solve tasks, with Elo ratings derived from blind pairwise comparisons.APEX-Agents-AA Benchmark LeaderboardArtificial Analysis' implementation of the APEX-Agents benchmark, testing AI agents on long-horizon, cross-application tasks in professional-services environments with realistic application tooling.𝜏²-Bench Telecom Benchmark LeaderboardA dual-control conversational AI benchmark simulating technical support scenarios where both agent and user must coordinate actions to resolve telecom service issues.Terminal-Bench Hard Benchmark LeaderboardAn agentic benchmark evaluating AI capabilities in terminal environments through software engineering, system administration, and data processing tasks.SciCode Benchmark LeaderboardA scientist-curated coding benchmark featuring 288 test set subproblems from 80 laboratory problems across 16 scientific disciplines.Artificial Analysis Long Context Reasoning Benchmark LeaderboardA challenging benchmark measuring language models' ability to extract, reason about, and synthesize information from long-form documents ranging from 10k to 100k tokens (measured using the cl100k_base tokenizer).AA-Omniscience: Knowledge and Hallucination BenchmarkA benchmark measuring factual recall and hallucination across various economically relevant domains.IFBench Benchmark LeaderboardA benchmark evaluating precise instruction-following generalization on 58 diverse, verifiable out-of-domain constraints that test models' ability to follow specific output requirements.Humanity's Last Exam Benchmark LeaderboardA frontier-level benchmark with 2,500 expert-vetted questions across mathematics, sciences, and humanities, designed to be the final closed-ended academic evaluation.GPQA Diamond Benchmark LeaderboardThe most challenging 198 questions from GPQA, where PhD experts achieve 65% accuracy but skilled non-experts only reach 34% despite web access.CritPt Benchmark LeaderboardA benchmark designed to test LLMs on research-level physics reasoning tasks, featuring 71 composite research challenges.ITBench-AA Benchmark LeaderboardArtificial Analysis' implementation of IBM's ITBench benchmark, testing AI agents on Kubernetes incident root-cause analysis from offline incident snapshots. The agent inspects alerts, events, traces, and topology and identifies the contributing-factor entities (deployments, pods, namespaces, network policies, etc.) responsible for the failure.Artificial Analysis Openness IndexA composite measure providing an industry standard to communicate model openness for users and developers.MMLU-Pro Benchmark LeaderboardAn enhanced version of MMLU with 12,000 graduate-level questions across 14 subject areas, featuring ten answer options and deeper reasoning requirements.Global-MMLU-Lite Benchmark LeaderboardA lightweight, multilingual version of MMLU, designed to evaluate knowledge and reasoning skills across a diverse range of languages and cultural contexts.LiveCodeBench Benchmark LeaderboardA contamination-free coding benchmark that continuously harvests fresh competitive programming problems from LeetCode, AtCoder, and CodeForces, evaluating code generation, self-repair, and execution.MATH-500 Benchmark LeaderboardA 500-problem subset from the MATH dataset, featuring competition-level mathematics across six domains including algebra, geometry, and number theory.AIME 2025 Benchmark LeaderboardAll 30 problems from the 2025 American Invitational Mathematics Examination, testing olympiad-level mathematical reasoning with integer answers from 000-999.MMMU-Pro Benchmark LeaderboardAn enhanced MMMU benchmark that eliminates shortcuts and guessing strategies to more rigorously test multimodal models across 30 academic disciplines.