From the source
Terminal X's Eval Agent, running on OpenAI's GPT-6.1 Sol, agreed with a senior human reviewer 88.6% of the time across 500 scored judgments, against 86.4% for the next-best model.
- The company sells a Capital Memory operating system that encodes how a firm invests into agents it owns, deployed on-premise or in the firm's own cloud, to address what it calls memory debt when senior investors leave.
- GPT-6.1 Sol handles the stages needing judgment and a large context window, while GPT-6 Luna runs high-volume steps where token efficiency matters more than depth.
- The Eval Agent grades every agent output against a 54-criterion rubric, and each failure becomes a Memory that teaches the system more about how that firm works, according to co-founder and CTO Ken Lee.
- GPT-6.1 Sol caught about 25% more recommendations that strayed from a fund's mandate and ran the same workflows at roughly 30% lower cost per query than the next-best model, at similar speed.
- Terminal X says one private equity firm reported $2.4 million in annualized savings from automated due diligence and investment memo workflows, and that it has built 1,700 firm Memories and run more than 1 million custom workflows.
