Hotel management 

Hotel operations workforce management: Employee profiles, facilities work orders, inventory, badge lifecycle, certification compliance, shift scheduling, payroll & benefits.

128

Test cases

32

Agent tools

Domain agentic intelligence index

We test models on private, non-contaminated tasks.
Here's what we found.

Composite pass^5 score (%)
Last updated: May 19
Composite pass^5 score (%)
Last updated: May 19

Composite pass^5 score across Tool use evaluations (higher is better).
Error bars show 95% confidence intervals.

Scaling curves

K = 1…5 runs

pass^k — Consistency

% tasks passed in every one of k runs.

0%10%20%30%40%50%60%70%k=1k=2k=3k=4k=5
GPT-5.4
Sonnet 4.6
Gemini 3 Flash
Minimax
Kimi K2.5
0%10%20%30%40%50%60%70%k=1k=2k=3k=4k=5
GPT-5.4
Sonnet 4.6
Gemini 3 Flash
Minimax
Kimi K2.5

Task difficulty distribution

Tasks bucketed by aggregate success rate

Buckets show difficulty tiers based on aggregate of models results on the benchmarking subset.

100%

0 of 123 tasks (0%)

0

75%+

7 of 123 tasks (6%)

7

50%+

56 of 123 tasks (46%)

56

25%+

60 of 123 tasks (49%)

60

0%

0 of 123 tasks (0%)

0

Example task

User Request

Correct Agent Solution

What Is Tested

Trusted by Leading AI Teams

Bank - internal HR dataset available for purchase