Terminal X investment evaluations match human judgment 88.6% of the time
What happened
On October 20, OpenAI News reported that Terminal X's Eval Agent, built on GPT-6.1 Sol, agreed with senior human reviewers on 88.6% of 500 scoring judgments, 220 basis points above the next-best model. The agent scores against 54 criteria, raises detection of answers that fall outside a fund's investment mandate by about 25%, and cuts cost per query by about 30%. The report also says a private equity client saved $2.4 million annualized by automating due diligence and investment memo workflows.
Written by AI from the coverage · updated 2 hours ago
Coverage
Follow the reports to see the story from different sides.
- OpenAI NewsTerminal X matches senior investor judgment 88.6% with GPT-6.1
Terminal X 的 Eval Agent 基于 GPT-6.1 Sol，在 500 条评分判断中与资深人类评审的一致率达 88.6%，比次优模型高 220 个基点。该 Agent 按 54 项标准评分，对偏离基金投资授权范围的回答检出率提升约 25%，单次查询成本降低约 30%。一家私募股权客户借助自动化的尽职调查与投资备忘录流程，实现年化 240 万美元节省。
Heat over time
Not enough continuous observations to draw a trend yet.