TSNBench: Benchmarking LLM Proficiency in Time-Sensitive Networking

TL;DR

TSNBench is a new benchmark evaluating large language models' proficiency in safety-critical Time-Sensitive Networking, revealing significant gaps in open-ended delay computation tasks despite high MCQ accuracy.

Contribution

The paper introduces TSNBench, the first comprehensive benchmark for LLMs in TSN, including expert-validated MCQs and open-ended WCD tasks with verified ground truths.

Findings

LLMs achieve 67-95% accuracy on MCQs but perform poorly on WCD tasks.

GPT-5 has a 36.2% MAPE on CBS WCD prediction, the best among tested models.

Most models' WCD prediction errors exceed 80%, risking real-time safety violations.

Abstract

We present TSNBench, the first benchmark for evaluating large language model (LLM) proficiency in Time-Sensitive Networking (TSN), a suite of IEEE 802.1 standards for deterministic communication with bounded latency in safety-critical domains such as autonomous vehicles, aviation, defense, and industrial automation. While LLMs have been extensively evaluated on general knowledge tasks, their capabilities in safety-critical networking domains remain largely unexplored. TSNBench comprises 939 expert-validated multiple-choice questions (MCQs) covering diverse TSN mechanisms, along with 100 open-ended Worst-Case Delay (WCD) computation tasks for Credit-Based Shaper (CBS) and Cyclic Queuing and Forwarding (CQF) across varying network topologies and traffic conditions. MCQ answers are validated by domain experts, and open-ended ground truth WCD values are computed using a verified Network Calculus (NC) solver for CBS and closed-form mathematical upper bounds for CQF. We evaluate 16 LLMs and find that although models achieve 67 to 95% accuracy on MCQs, they fail substantially on open-ended WCD computation. For CBS, only GPT-5 achieves a Mean Absolute Percentage Error (MAPE) of 36.2%, meaning its predicted WCD deviates by 36.2% of the actual TSN flow delay on average, while most models exceed 80%. For CQF, the best model achieves 41.8% MAPE, with most models clustering between 80% and 100%. Such errors are large relative to TSN latency budgets and can lead to violations of real-time constraints and unsafe configurations. TSNBench demonstrates that MCQ benchmarks may overestimate LLM capabilities in safety-critical networking domains.