Terminal X says GPT-6.1 matches senior investors on 88.6% of judgments
Original title: Terminal X matches senior investor judgment 88.6% with GPT-6.1
Across 500 scored judgments it beat the next model by 220 basis points and cut query cost about 30%.
OpenAI published a customer story on Terminal X’s Capital Memory system, which encodes a firm’s investment process into agents the firm owns, on-premise or in its own cloud. GPT-6.1 Sol handles long-context judgment stages, while GPT-6 Luna runs high-volume steps. Terminal X says an Eval Agent on a 54-criterion rubric agreed with a senior reviewer 88.6% of the time across 500 judgments, versus 86.4% for the next-best model, and ran the same workflows at about 30% lower cost per query. Savings claims, including $2.4 million annualized at one private equity firm, come from Terminal X.
Unconfirmed: Agreement, cost and savings figures are Terminal X’s own results, not an independent benchmark.
- Terminal X reports 1,700 firm Memories and more than 1 million custom workflows.
- GPT-6.1 Sol caught about 25% more mandate departures than the next-best model.
- Agents process English, Korean, Japanese and Chinese documents in a private data room.
- Weekly active users equal 81% of monthly active users, the company says.
- Vendors
- OpenAI
- Products
- Capital Memory
- Organizations
- Terminal X
- AI models
- GPT-6.1 SolGPT-6 Luna
From the original article
4 min read at openai.com
Terminal X’s Eval Agent grades outputs against a 54-criterion rubric, catching 25% more deviances from a fund’s mandate with GPT-6. 1 Sol.
openai.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.