APEX-Agents · Law
World425_amk_01
APEX-Agents task World425_amk_01 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
Summit filed Form 2553 (Rev B) with the Internal Revenue Service ("IRS") on February 19, 2019. Do you see any potential problems under Treasury Regulation 1.1362-6(b)? Give me your analysis back here so I can decide what to do.
Published trajectories
Agent runs on this task
Curated dual-harness runs (parsed + original sandbox). Best scored run per model.
| Model | Harness | Score | Result | Links |
|---|---|---|---|---|
| Gemini 3 Flash | dual | 6/6 | Pass | Run detailsPublic trace |
| Gemini 3.1 Pro | dual | 6/6 | Pass | Run detailsPublic trace |
| GPT-5.5 | dual | 6/6 | Pass | Run detailsPublic trace |
| GPT-5.4 | dual | 1/6 | Fail | Run detailsPublic trace |
| GPT-5.4 mini | dual | 1/6 | Fail | Run detailsPublic trace |
| GPT-5.4 nano | dual | 1/6 | Fail | Run detailsPublic trace |
| fireworks models Kimi K2 | 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 that Treasury Regulation 1.1362-6(b)(2)(i) requires that when the stock of the corporation is owned by husband and wife as community property, each spouse must consent to the election
States that Wisconsin is a community property state
States that IRS Form 2553 provides that shareholder Gregory Beaumont resides in a community property state
States that Gregory Beaumont's marital status is unclear as of February 19, 2019
States that if Gregory Beaumont was married as of February 19, 2019, a consent from his spouse is potentially missing
States that Gregory Beaumont's marital status at the time of signing Form 2553, February 19, 2019 should be confirmed