Free Intensive Nov 21–22Free 2-day hands-on Vibecoding Intensive on Nov 21–22 at 16:00 MSK (Sat–Sun) · Live Cohort 3 starts Nov 23Join Free →Cohort 3 (Nov 23)
IndustryInsight · On RadarTechCrunch→ original

Model ML Raises $250 Million for Banker-Focused AI Agents at a $1.5 Billion Valuation

London-based Model ML has raised $250 million at a $1.5 billion valuation for AI agents that handle junior bankers' routine work: financial models, pitch decks and due diligence reports. The agents work inside Excel and PowerPoint and update spreadsheets and slides on their own when new data appears in a deal. According to the company, open models cost roughly 2% of the price of top models. Takeaway: a narrow agent built for a single profession can attract big checks.

AI-processed from TechCrunch; edited by Hamidun News
Model ML Raises $250 Million for Banker-Focused AI Agents at a $1.5 Billion Valuation
Source: TechCrunch. Collage: Hamidun News.
Listen to article
Article Fact Sheet & Key Figures
ParameterDetails
Primary sourceTechCrunch ↗
Editor-in-Chief
Key metric #1The Series B was led by General Atlantic and QED Investors
Key metric #2A year earlier, the startup raised $75 million in a Series A
Key metric #3Competitor Rogo raised $160 million in April at a $2 billion valuation

A large share of a junior investment banker's work is assembly; analysis takes up less of the time. Recalculate a model after yet another edit, move the numbers into a presentation, boil dozens of documents down into a single report. London startup Model ML wants to take that routine off people's hands, and it has just raised $250 million in a Series B at a $1.5 billion valuation.

What Model ML sells

The company builds AI agents for investment bankers and consultants. The agents take over tasks that traditionally fall to juniors: building financial models, preparing pitch decks (presentations for clients and investors) and writing due diligence reports, meaning checks on a company before a deal.

The agents are built into familiar tools. They connect to the client's systems, data and software, including Excel and PowerPoint, where bankers already spend most of their day. Think of an assistant that tracks the whole deal. When something new comes up, such as changed terms or fresh figures, it updates the financial model on its own and rebuilds the slides that depend on it. According to the company, the agents understand the full context of the deal that bankers and consultants are working on.

An ordinary chatbot is a poor fit for this kind of task: an edit to one cell triggers a dozen changes across spreadsheets and the presentation, and a person spends hours making sure nothing is missed.

Money, investors and team

The Series B came a year after the $75 million Series A. It was led by General Atlantic and existing investor QED Investors. The startup is three years old. It was founded by brothers Chaz and Arni Englander, who have built companies before. Their past projects include Fat Llama, a peer-to-peer rental marketplace. Sweden's Hygglo bought it in 2022 for $41.5 million.

Model ML is not alone in this market. Hebbia and Rogo are also automating the routine work of bankers and management consultants; Rogo raised $160 million in April at a $2 billion valuation. Anthropic and OpenAI have also released tools for finance professionals. So Model ML competes both with niche startups and with the labs that build the most powerful models.

A bet on open models

Model ML's main argument against the bigger players is cost. The company says its agents run on open models, so they cost roughly 2% of the price of frontier closed models and deliver results faster. When a single report or deal model requires many calls to a neural network, a price gap like that translates directly into margin.

An important caveat: both the 2% figure and the speed claim come from the company itself. The announcement offers no independent verification. It also gives no data on revenue, number of customers, or how the agents handle errors in financial models, where inaccuracy is costly. It does not say which open models are used, either.

What this means for practitioners

For founders, the story shows that investors are willing to write big checks for vertical agents built into the working tools of a specific profession. Excel and PowerPoint remain the place where users already live, and integrating with them removes a barrier to adoption.

For developers and product managers, the lesson is different: in a narrow, well-defined task, open models can compete with frontier ones on price and speed. What separates a full product from a wrapper around someone else's API is increasingly the depth of integration and the understanding of the work context. If you are building an agent for a profession with a lot of repetitive routine, it is worth calculating what share of tasks you can hand off to a cheaper model.

Model ML's claims can be tested by its customers, meaning banks and consulting firms. The $250 million round itself shows that investors believe in this bet.

Story

Key Questions Answered

What Model ML sells?

The company builds AI agents for investment bankers and consultants. The agents take over tasks that traditionally fall to juniors: building financial models, preparing pitch decks (presentations for clients and investors) and writing due diligence reports, meaning checks on a company before a deal. The agents are built into familiar tools. They connect to the client's systems, data and software, including Excel and PowerPoint, where bankers already spend most of their day.

What are the key details on Money, investors and team?

The Series B came a year after the $75 million Series A. It was led by General Atlantic and existing investor QED Investors. The startup is three years old. It was founded by brothers Chaz and Arni Englander, who have built companies before. Their past projects include Fat Llama, a peer-to-peer rental marketplace. Sweden's Hygglo bought it in 2022 for $41.5 million. Model ML is not alone in this market.

What are the key details on A bet on open models?

Model ML's main argument against the bigger players is cost. The company says its agents run on open models, so they cost roughly 2% of the price of frontier closed models and deliver results faster. When a single report or deal model requires many calls to a neural network, a price gap like that translates directly into margin. An important caveat: both the 2% figure and the speed claim come from the company itself. The announcement offers no independent verification.

ZK
Zhemal Khamidun — CPO at Alpina Digital, product manager of AlpinaGPT. Building AI products since 2019. Hamidun News is a pet project that grew into a production aggregator with 6,000+ readers.
✦ Private AI Builders Community · Vibe Club

Ship AI Faster with Founders, Engineers & Weekly Architecture Reviews

Inside Vibe Club (hamidun.com/club) — weekly live masterminds with Zhemal Khamidun, Tech Radar, full LMS access, and 370+ production Claude Code skills.

Join Vibe Club →Explore Hamidun Academy
What do you think?
Loading comments…