APEX-Agents · Investment Banking

World228_JP_01

0/1Fail

APEX-Agents task World228_JP_01 in AI Agents for Merger Models. Compare dual-harness agent runs across models — rubric criteria, scores, and public traces.

AI Agents for Merger ModelsInvestment Banking World 228Dual harnessGrader: rubric

Task IDtask_326a170f1d6440988779b2ef1dac919f

WorldInvestment Banking World 228

Expected outputmessage_in_console

Published runs4 models · dual config

Task prompt

What the agent was asked to do

Using the 2024 ATR annual report, reply to me with the absolute performance difference between ATR shares and the Peer Group in 2024 as a percentage (round this to two 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-5.5showcase	dual	0/1	Fail	Share page Public trace
Gemini 3.1 Pro	dual	1/1	Pass	Share page Public trace
GPT-5.4 mini	dual	0/1	Fail	Share page Public trace
GPT-5.4 nano	dual	0/1	Fail	Share page Public trace

Grading rubric

Criteria and grader verdict (showcase run)

States the difference in share performance is 28.61%
Fail
Evidence: <TEXT_RESPONSE> states “`|144.64 - 108.24| = 36.40%`” and “Answer: 36.40%.” Assessment: The criterion requires that the response “States the difference in share performance is 28.61%.” The response gives 36.40%, not 28.61%, so it fails.