Model ML raises $250 million to automate the work junior bankers do

4 Sources

Share

Model ML, a three-year-old London startup building autonomous AI agents for investment bankers, consultants and investors, has raised $250 million in a round valuing it at $1.5 billion

1

3

2

.

Source: PYMNTS

Source: PYMNTS

Model ML's new round was co-led by General Atlantic and QED Investors, taking its total funding to $350 million

1

2

. Evolution Equity Partners, Airtree, Forestay Capital, Y Combinator, FT Partners, Latitude, Inception Fund and 13Books Capital also took part

3

2

.

What the agents do

The company's software runs on a client's own systems and data, plugging into tools such as Excel and PowerPoint

1

3

2

. It produces financial models, pitch decks and due diligence reports, work traditionally done by junior bankers

1

3

2

. Model ML says its agents act on their own: they anticipate what matters, judge whether breaking news on a covered company is significant, and turn out analysis in whatever format the firm uses, in minutes instead of days

4

3

.

The growth figures investors are buying are the company's own reported numbers

3

2

. Model ML says that in Q3 2026 annual recurring revenue grew twelvefold year on year, with daily active users up twenty-sixfold

3

2

. The new round lands about a year after a $75 million Series A led by FT Partners, with HSBC Asset Management investing along the way in August

2

. The company says the money will fund research and engineering hiring in London and New York, which it frames as deepening product development while supporting wider customer deployments

4

3

2

.

Cheaper models, crowded field

Underneath the product is a bet on open-source models. Model ML benchmarks them against real financial work, then routes each task to whichever model handles it best, rather than tie itself to one AI provider

1

3

2

. The company says steering away from any single AI provider is what lets it cut costs while improving quality and speed

1

3

. By its account, the models it runs cost about 2% of what frontier models do and, in most cases, work five times faster on the jobs they handle

1

3

2

. It also says banks' customer data stays out of AI model training, an assurance aimed at firms handling sensitive client information

2

.

The field is well funded. New York rival Rogo raised $160 million at a $2 billion valuation back in April, Hebbia operates in the same space, and OpenAI and Anthropic sell AI tools to financial professionals directly

1

2

.

Revenue, not redundancies

Brothers Chaz and Arnie Englander founded the company in 2023 after co-founding Fat Llama, a peer-to-peer rental marketplace that Sweden's Hygglo bought for $41.5 million in 2022

1

3

2

. This year Model ML opened new offices in Paris, Madrid, Tokyo, Singapore and Dubai

4

3

.

Source: Finextra Research

Source: Finextra Research

Englander's point is that the tools have moved past saving time. The average user is 33, not a fresh graduate, and the software brings in revenue instead of trimming headcount. Englander said revenue comes in roughly equal shares from banks, consulting firms and asset managers

2

. That framing lands against a backdrop of banks linking layoffs to AI

2

. Barclays said in October it would cut 5,000 roles, and Standard Chartered plans 7,800 back-office cuts by 2030

2

. Englander told Bloomberg the argument in his own words: "The average age of a Model ML user is 33 years old, not 23 years old. These tools are no longer productivity and efficiency tools. They're revenue generative."

© 2026 TheOutpost.AI All rights reserved