APEX-Agents · Investment Banking
World225_RL_Task01
APEX-Agents task World225_RL_Task01 in AI Agents for Tax Due Diligence. Compare dual-harness agent runs across models, scores, and public traces.
Task prompt
What the agent was asked to do
In the existing sheet for the REIT model, create a sensitivity table for 2029E Net Income. Show Revenue Growth Step-up at 50 basis points and 100 basis points. Show EBITDA Margin Step-up at 20 basis points and 40 basis points. Round all final numbers to the nearest million. To calculate the correct values, assume that revenue growth in 2025E starts at 8.0% and increases by the same amount every year of the projection period, referred to as the "Revenue Growth Step-up". Assume that EBITDA Margin % starts at 44.0% and increases by the same amount every year of the projection period, referred to as the "EBITDA Margin Step-up".
Published trajectories
Agent runs on this task
Curated dual-harness runs (parsed + original sandbox). Best scored run per model.
| Model | Harness | Score | Result | Links |
|---|---|---|---|---|
| GPT-5.4 mini | dual | 4/4 | Pass | Run detailsPublic trace |
| GPT-5.5 | dual | 4/4 | Pass | Run detailsPublic trace |
| Gemini 3.1 Pro | dual | 0/4 | Fail | Run detailsPublic trace |
| GPT-5.4 nano | dual | 0/4 | Fail | Run detailsPublic trace |
Grading rubric
Rubric criteria
Runs are graded against these criteria. Open a run for model-specific verdicts.
States 2029E Net Income in the scenario with 50 basis points of Revenue Growth Step-up and 20 basis points of EBITDA Margin step-up is 1,772 million
States 2029E Net Income in the scenario with 100 basis points of Revenue Growth Step-up and 20 basis points of EBITDA Margin step-up is 1,887 million
States 2029E Net Income in the scenario with 50 basis points of Revenue Growth Step-up and 40 basis points of EBITDA Margin step-up is 1,817 million
States 2029E Net Income in the scenario with 100 basis points of Revenue Growth Step-up and 40 basis points of EBITDA Margin step-up is 1,933 million