All evaluations

๐œยฒ-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

๐œยฒ-Bench (Tau-2 Bench) introduces a new paradigm for evaluating conversational AI by simulating both the agent and user to actively modify a shared world state.
The telecom domain tests agents' abilities to guide users through technical troubleshooting to test problem-solving and effective communication skills.
Developed by Sierra Research, this benchmark addresses gaps between other benchmarks and real-world customer service scenarios where users are active participants in problem resolution.

All evaluations are conducted independently by Artificial Analysis. More information can be found on our Intelligence Benchmarking Methodology page.

Publication

View on arXiv

๐œยฒ-Bench: Evaluating Conversational Agents in a Dual-Control Environment

Victor Barres, Honghua Dong, Soham Ray, Xujie Si, Karthik Narasimhan.

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 ๐œยฒ-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, ๐œยฒ-bench provides a controlled testbed for agents that must both reason effectively and guide user actions.

๐œยฒ-Bench Telecom

JT-35B-Flash scores the highest on ๐œยฒ-Bench Telecom with a score of 99.1%, followed by GLM-5.2 (max) with a score of 99.1%, and GLM-4.7-Flash (Reasoning) with a score of 98.8%

Score

๐œยฒ-Bench Telecom Benchmark Leaderboard: Score

Independently benchmarked by Artificial Analysis
Reasoning models are indicated by a lightbulb icon

Token Usage

๐œยฒ-Bench Telecom Benchmark Leaderboard: Output Tokens per Task

Output tokens used to run one task, broken down by reasoning and answer tokens
Reasoning models are indicated by a lightbulb icon

The average number of answer and reasoning tokens produced per benchmark task in this evaluation.

CostUpdated

๐œยฒ-Bench Telecom Benchmark Leaderboard: Cost per Task

Average cost per task (USD), broken down by input, cache hit, cache write, reasoning, and answer tokens
Reasoning models are indicated by a lightbulb icon

Average cost per task in the evaluation. Costs are split by input, cache hit, cache write, reasoning, and answer token pricing where canonical token counts are available.

SpeedUpdated

๐œยฒ-Bench Telecom Benchmark Leaderboard: Time per Task

Weighted average wall clock time (minutes) per task; excludes TTFT and execution time ยท Lower is better
Reasoning models are indicated by a lightbulb icon

The weighted average time (seconds) per evaluation task. This is calculated by dividing output tokens per task by output speed, weighted by the relative weights of each benchmark in the evaluation.

Score vs. Release Date

๐œยฒ-Bench Telecom Benchmark Leaderboard: Score vs. Release Date

Most attractive region

Example Tasks

Explore Evaluations

Artificial Analysis Intelligence IndexArtificial Analysis Intelligence Index

A composite benchmark aggregating nine challenging evaluations to provide a holistic measure of AI capabilities across mathematics, science, coding, and reasoning.

GDPval-AA v2 LeaderboardGDPval-AA v2 Leaderboard

GDPval-AA v2 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 LeaderboardAPEX-Agents-AA Benchmark Leaderboard

Artificial Analysis' implementation of the APEX-Agents benchmark, testing AI agents on long-horizon, cross-application tasks in professional-services environments with realistic application tooling.

๐œยฒ-Bench Telecom Benchmark Leaderboard๐œยฒ-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.

๐œยณ-Banking Benchmark Leaderboard๐œยณ-Banking Benchmark Leaderboard

A fintech customer-support benchmark from the ๐œ-Knowledge framework that tests whether agents can navigate a large unstructured knowledge base and execute multi-step tool calls to resolve realistic banking workflows.

Terminal-Bench Hard Benchmark LeaderboardTerminal-Bench Hard Benchmark Leaderboard

An agentic benchmark evaluating AI capabilities in terminal environments through software engineering, system administration, and data processing tasks.

Terminal-Bench v2.1 Benchmark LeaderboardTerminal-Bench v2.1 Benchmark Leaderboard

A verified refresh of Terminal-Bench v2.0 โ€” 89 curated tasks across software engineering, system administration, data processing, model training, and security, with environment and instruction fixes so scores reflect agent capability rather than environment gaps.

SciCode Benchmark LeaderboardSciCode Benchmark Leaderboard

A scientist-curated coding benchmark featuring 288 test set subproblems from 80 laboratory problems across 16 scientific disciplines.

Artificial Analysis Long Context Reasoning Benchmark LeaderboardArtificial Analysis Long Context Reasoning Benchmark Leaderboard

A 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 BenchmarkAA-Omniscience: Knowledge and Hallucination Benchmark

A benchmark measuring factual recall and hallucination across various economically relevant domains.

IFBench Benchmark LeaderboardIFBench Benchmark Leaderboard

A 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 LeaderboardHumanity's Last Exam Benchmark Leaderboard

A 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 Leaderboard

The most challenging 198 questions from GPQA, where PhD experts achieve 65% accuracy but skilled non-experts only reach 34% despite web access.

CritPt Benchmark LeaderboardCritPt Benchmark Leaderboard

A benchmark designed to test LLMs on research-level physics reasoning tasks, featuring 71 composite research challenges.

ITBench-AA Benchmark LeaderboardITBench-AA Benchmark Leaderboard

Artificial 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 IndexArtificial Analysis Openness Index

A composite measure providing an industry standard to communicate model openness for users and developers.

MMLU-Pro Benchmark LeaderboardMMLU-Pro Benchmark Leaderboard

An 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 LeaderboardGlobal-MMLU-Lite Benchmark Leaderboard

A lightweight, multilingual version of MMLU, designed to evaluate knowledge and reasoning skills across a diverse range of languages and cultural contexts.

LiveCodeBench Benchmark LeaderboardLiveCodeBench Benchmark Leaderboard

A 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 LeaderboardMATH-500 Benchmark Leaderboard

A 500-problem subset from the MATH dataset, featuring competition-level mathematics across six domains including algebra, geometry, and number theory.

AIME 2025 Benchmark LeaderboardAIME 2025 Benchmark Leaderboard

All 30 problems from the 2025 American Invitational Mathematics Examination, testing olympiad-level mathematical reasoning with integer answers from 000-999.

MMMU-Pro Benchmark LeaderboardMMMU-Pro Benchmark Leaderboard

An enhanced MMMU benchmark that eliminates shortcuts and guessing strategies to more rigorously test multimodal models across 30 academic disciplines.