ITSMBench C
AI benchmark · Agentic
What is ITSMBench?
ITSMBench evaluates whether AI agents can resolve real enterprise IT service management tickets end-to-end. Every task runs inside a simulated enterprise of 42 mocked software applications (Okta, Jira Service Management, Slack, PagerDuty, ServiceNow, Intune, and more) backed by roughly 1,800 database tables and 2,000+ REST endpoints. It scores 89 expert-written tasks spanning L2 service management and L3 IT operations across five ops roles (Device, Network, Access, SecOps, Compliance), each graded by 10-30 deterministic verifiers over the final environment state — no LLM judge. Agents must discover the right tools, trace the actual root cause, follow the runbook, and leave every touched system consistent. The public leaderboard reports Pass@1 with standard error, average cost per task, and output tokens for each model, harness, and reasoning-effort configuration. Built by Atomicwork in collaboration with New Measure; tasks and environments are open-sourced.
Tier C
Niche: a handful of models scored, or built by an RL vendor.
How much stands behind a tier: usage, who backs it, a confirmed line from its own site or paper, and a clear read on X. High with three of the four, medium with two, low otherwise.
- 5 models scored
- in curated RL directories (1 pt)
How we got to tier C
- Adoption: 5 models scored. Tier C.
- Not yet read on X.
- In curated RL directories: 1 point, a floor of none.
- Not yet confirmed by a line from its own site or paper.
A clear view on X moves a tier one step up or down; most entries don't move. How tiers work.
Not on the map yet: no line from its own site or paper places it in a domain.
Next: browse all benchmarks. Entry from the RL Research daily scrape of public sources, 2026-10-11 snapshot.