Deep dives into private markets due diligence, AI benchmarks, and operational efficiency
ToltIQ's latest research explores the strengths and limitations of standalone Large Language Models (LLMs) like ChatGPT and Claude compared to our integrated platform. Our findings reveal a clear advantage: ToltIQ delivers superior accuracy, relevance, and thoroughness, enabling businesses to access comprehensive insights...
Private equity firms have increasingly adopted AI tools for due diligence, but rigorous, use case-specific benchmarks have been largely absent from the literature. This study addresses that gap.
Family offices have spent the last decade building serious private markets capabilities. But moving from fund commitments to co-investments and direct deals changes the job entirely. Here's why the usual diligence solutions fall short — and what purpose-built infrastructure actually enables a lean team to do.
The secondaries market is scaling faster than traditional diligence can support. This article examines how investors are reviewing more data under tighter timelines and evolving confirmatory diligence to move faster while making decisions with greater confidence.
Human capital was once the primary edge in PE due diligence. Today, that edge increasingly comes from how well firms harness AI. Client Director Eric Jacobs, who is on the front lines helping dozens of firms implement AI workflows, shares five ways AI is already reshaping private equity and changing how firms work.
The institutional knowledge your deal team built during diligence shouldn't disappear the moment the transaction closes. When each stage of a deal builds on what came before and informs what comes next, value compounds instead of disappearing.
Private equity operations are fundamentally changing as teams integrate artificial intelligence technologies to enhance data analysis capabilities, streamline deal sourcing, and improve portfolio management tools. To maximize these benefits, firms must develop specific implementation strategies such as...
Oil and gas M&A is not becoming simpler. That means finding ways to reduce manual synthesis, surface risks earlier, and maintain alignment across workstreams throughout the deal process.The opportunity is to bring diligence workstreams, data sources, and risk insights into a unified view.
Poor prompting is the silent productivity killer in PE diligence, and most don’t even realize it’s happening. Through our work with thousands of users, we've identified the five most common patterns that limit prompt quality. For each pattern, we'll show why it fails and how to fix it, with before-and-after examples that demonstrate the difference.
It's a question we hear frequently from private equity professionals evaluating AI tools for deal work. And it's a fair question. ChatGPT and Claude are remarkably powerful foundation models that have transformed how people interact with documents and data. These same models underlie ToltIQ's platform. Here's what makes the difference.
When you're reviewing 80 opportunities to find one investment, the value of thinking faster and better about which deals deserve your attention compounds quickly. It's about decision quality at the point where your time investment is still minimal.
In 2025, 86% of private equity firms indicated they’ve introduced AI into their workflows but only around 20% have really operationalized use cases with concrete results. The gap isn't about AI technology capability. It's about understanding the actual problem they are trying to solve, which invariably involves documents more than structured data.
This study is a comparative evaluation of the currently released versions of three leading large language models (LLMs), GPT 4.1, Claude Sonnet 4, Gemini 2.5 Pro Preview, and their performance on several financial due diligence tasks relevant to private equity. Utilizing ToltIQ’s database of hundreds of real-world use cases...
Private markets professionals navigate complexity that compounds across sectors: each segment within private markets, whether it is infrastructure, real estate, natural resources, or technology, operates with its own distinct documentary framework. Each document tells part of the business story, but the real insight comes from...
This report highlights the distinctive traits of leading AI models utilized at ToltIQ. The models included in this benchmark were ChatGPT 5, ChatGPT 4.1, Gemini 2.5 Pro, and Claude 4 Sonnet. We analyzed their performance when leveraged in the ToltIQ platform and scored their responses based on six different evaluation criteria: accuracy, prompt...
In the race to automate AI evaluation, we risk losing sight of what matters most: the human experience of AI output. At ToltIQ, where we've conducted systematic benchmarking across all of the leading AI models, we've learned that valuable insights can’t come from automated metrics alone—they emerge from the careful intersection of human judgment...
ToltIQ's comprehensive prompt engineering evaluation tested hundreds of due diligence queries, using various prompting approaches, to identify which approaches deliver the best results across reasoning, relevance, problem solving, and accuracy. Our analysis revealed critical patterns that dramatically impact AI performance.
In the ever-evolving world of data analytics, the ability to create effective and insightful visualizations is crucial. At ToltIQ, we leverage the Assistant API from OpenAI to enhance our Data Chat capabilities, making it easier than ever to generate high-quality charts and visualizations from datasets in xlxs files.
At ToltIQ, we are continually pushing the boundaries of data visualization and chart analysis through the integration of Generative AI. Spearheaded by Software Engineer, Michael Scoleri, our recent efforts have focused on addressing the challenges of accurately reading and interpreting charts using Large Language Models.
As we ramp up our pilot clients at ToltIQ, we are learning a lot about the art of constructing system prompts, prompt engineering to get at the right data, how 3.5 vs 4 behave differently, what temperature adjusts in a conversation and the different ways of embedding, chunking and meta tagging documents to discover...
The relentless stream of new language models hitting the market presents both opportunities and challenges for enterprises leveraging AI. Technical teams face integration hurdles while compliance officers worry about the governance implications of model changes. At ToltIQ, where private equity clients leverage...
In the rapidly evolving landscape of AI-powered solutions, there's a tendency to build platforms that promise to solve every problem across every business area. At ToltIQ, we've taken a different approach. We're laser-focused on one critical area: post-signing, VDR due diligence in private markets through specialized generative AI.
The financial services industry has spent decades perfecting data infrastructure for public markets. Clean, standardized feeds power everything from trading to risk management systems. So when private markets firms look to modernize their data capabilities, the natural instinct for many firms is to try and replicate what works in public markets...
Here we go again … Gemini is actually multiple products (similar to OpenAI’s GPT). It will initially be the “new” back end to Google Bard for consumers and the API is supposed to be available to developers and enterprise customers in early 2024. It’s designed to be multi-modal from the start. Things are going to get interesting...
As if GPT4 API wasn’t enough to change the landscape, OpenAI is now rolling out Code Interpreter to all paid ChatGPT Plus customers over the next week (it will not available via the API). I’ve got it and it’s very cool. I feel like a kid in a candy store with $20 circa 1980s! It’s literally a magic box that allows you to upload data files...
Q: How are advanced AI tools changing the due diligence landscape for investment teams? RS: The private markets industry is at an inflection point. It’s clear that AI is enhancing due diligence and in the process it is also fundamentally redefining it. What once took weeks now takes hours. Insights that might have been missed entirely....
Private markets infrastructure investors play a vital role in advancing long-term, capital-intensive projects that underpin economic growth and societal resilience. These include energy infrastructure, data centers, transportation systems, and other critical assets that communities rely on for decades. Embedded in this mission...
2023 was a wild year in the world of Generative AI. It also has so much potential to dramatically impact business workflows (not to mention personal creativity) that with a strong nudge from Matthew Brandman, I came out of retirement and started ToltIQ. We put together a team with deep PE and Credit industry knowledge...
ToltIQ's new Agent Mode is our latest step in making Large Language Models (LLMs) work better for you. The Agent Mode feature adds an intelligent decision-making layer to query processing. At its core, Agent Mode acts as an intermediary, refining user queries and determining the most effective path for information retrieval and analysis.
If you ask an investment professional to list the apps they use daily for financial analysis, it’s likely limited to Excel, Adobe Acrobat, and a handful of financial data providers. These apps share a common trait: tools so ubiquitous they can get away with being unintuitive—how many apps can force users to learn Alt-HOI?
In the expanding world of artificial intelligence, prompt engineering has become an important tool to obtain the best output. Prompt engineering is the process of optimizing prompts to improve the quality, accuracy, and relevance of responses. Large language models (LLMs) are sensitive to minor changes in prompts, proving that prompt...
This study presents a comparative evaluation of two large language models LLMs), Claude 3.7 Sonnet and Claude 4 Sonnet, in the context of private equity due diligence applications. Through rigorous benchmarking across hundreds of real-world use cases, we assessed model performance across multiple dimensions: reasoning, relevance, speed...
Stay updated with the latest in AI through this curated collection of articles, research, and innovations. Highlights include Andrew Ng’s course on advanced retrieval, emerging AI research themes, Anthropic’s legal protections, OpenAI’s prompt engineering guide, and DeepMind’s FunSearch. Discover AI's role in healthcare, legal models, video search
In the fast-paced world of AI, it's crucial to grasp how our tools are evolving. Last month AI researchers, Lila Karl and Michael Van Demark, conducted a comprehensive analysis comparing the May and August versions of the ChatGPT-4 Omni model. Their findings reveal both significant improvements and ongoing challenges that are crucial...
Here are some cool custom instructions if you’re playing with chatGPT. This provides grounded and deterministic answers from the LLM. Copy the section after “What would you like chatGPT to know” into the top custom instruction and then the bottom section starts after “How would you like chatGPT to respond”...
This guidance from Anthropic for their top end model is fascinating: Adding the sentence “Here is the most relevant sentence in the context:” to the start of Claude’s response was enough to raise Claude 2.1’s score from 27% to 98% on the original evaluation. Anthropic’s Claude is one of the major foundation LLMs alongside OpenAI’s GPT4.
At ToltIQ, we're witnessing significant improvements in model capabilities and they are happening faster than expected, yet again! Over the past couple of weeks, the performance of the three primary models we rely on - GPT-4o, Claude 3.7 Sonnet with reasoning enabled, and Gemini 2.5 - have markedly improved when accessed via API.
