APEX-Agents · Investment Banking
World221_TR_08
APEX-Agents task World221_TR_08 in AI Agents for SEC Disclosure Analysis. Compare dual-harness agent runs across models, scores, and public traces.
Task prompt
What the agent was asked to do
Use "Comp Analysis – Modify" as a clean version of Comp Analysis. Replace peers WhiteHorse Finance (WHF) and TriplePoint Venture Growth BDC corps. (TPVG) and replace with SLR Investment Corp. (SLRC), Gladstone Capital (GLAD) and Stellus Capital Investment Corp (SCM). Use the attached SEC filings to complete the task. Retrieve the AUM, NAV / Share and Debt / Equity as of September 30, 2025, to complete the analysis. Calculate the mean and median for each key metrics considering BDCs with a Market Cap in the $350M to $900M range only. Get your answer back to me right here. Present the AUM in $M. All the final numbers must be rounded to 2 decimal places.
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 | 4/6 | Fail | Run detailsPublic trace |
| GPT-5.4 nano | dual | 3/6 | Fail | Run detailsPublic trace |
| Gemini 3.1 Pro | dual | 2/6 | Fail | Run detailsPublic trace |
| GPT-5.5 | dual | 0/6 | Fail | Run detailsPublic trace |
Grading rubric
Rubric criteria
Runs are graded against these criteria. Open a run for model-specific verdicts.
States the Mean AUM is $2,635.91 million
States the Mean NAV / Share is $17.27
States the Mean Debt / Equity is 1.22x
States the Median AUM is $2,400.00 million
States the Median NAV / Share is $17.40
States the Median Debt / Equity is 1.28x