Ramp, the corporate-card and expense-management company, told investors on October 15, 2026 that it had closed a $1.85 billion primary financing round at a $60 billion valuation. The round, led by Dragoneer Investment Group, lands just four months after Ramp raised $750 million at a $44 billion valuation in June 2026, a jump that puts Ramp among the fastest-climbing private companies in the current funding cycle. Co-CEO Eric Glyman’s company now carries a valuation that would rank it among the largest privately held technology firms in the United States, and the round’s framing leans heavily on artificial intelligence rather than on the corporate-card business that built Ramp in the first place.
The timing matters. Ramp’s pitch to investors has shifted from a straightforward spend-management story to one built around AI agents, developer tooling, and tokens, placing the company squarely inside the AI coding and agent tooling boom that has reshaped venture math across software this year. What follows is a breakdown of the round itself, what Ramp’s own technology choices say about where fintech is headed, and how the $60 billion figure stacks up against the other mega-valuations stacking up in 2026.
Ramp’s $60 Billion Round, By the Numbers
Ramp confirmed the financing on Thursday, describing it as a $1.85 billion primary round, meaning the new capital went into the company’s balance sheet rather than paying out existing shareholders. The round values Ramp at $60 billion, according to TechCrunch and corroborated by Bloomberg, which first reported the deal was taking shape in early October. Ramp said the financing brings its total equity raised to date to roughly $5 billion, a figure that spans every round since the company’s founding.
Alongside the valuation number, Ramp disclosed that its annualized revenue has now topped $2 billion, a milestone the company highlighted prominently in its own announcement. That revenue figure is the clearest signal yet that Ramp’s business, selling corporate cards and expense-management software to other companies, has scaled well beyond its early days as a Y Combinator-style expense tool. Investors appear to be pricing in both that revenue trajectory and the company’s expanding bet on AI products.
Who’s Writing the Checks: Dragoneer, Thrive, and a Long Investor List
Dragoneer Investment Group led the round, with Thrive Capital, Founders Fund, Sands Capital, and Avenir Growth named as co-leads. That lineup mixes a growth-equity specialist in Dragoneer with Thrive Capital, a firm whose bets on OpenAI and other AI-era companies have made it one of the more closely watched names in late-stage venture right now.
Returning backers
The round also drew a long list of participants with names familiar to anyone who tracks late-stage tech financing: Capital International, Fidelity Management & Research Company, T. Rowe Price, D1 Capital Partners, Greenoaks, ICONIQ, GIC, Coatue, D.E. Shaw, General Catalyst, Khosla Ventures, Lightspeed Venture Partners, and Stripes, among others. The presence of mutual-fund managers like Fidelity and T. Rowe Price alongside sovereign and pension capital such as GIC and Ontario Teachers’ Pension Plan suggests Ramp is being underwritten less like a typical venture-backed startup and more like a pre-IPO growth company.
A crowded cap table
Additional names on the round include 8VC, 1789 Capital, Alpha Wave Global, Definition Capital, FirstMark, Generation Investment Management, Glade Brook Capital Partners, Pinegrove Opportunity Partners, and Soma Capital. A roster that long is itself a data point: it signals strong demand to get allocation in a company that, four months earlier, was valued at roughly two-thirds of its current price.
From $44 Billion to $60 Billion in Four Months
The pace of Ramp’s valuation increases is the most striking part of this story. In June 2026, Ramp raised $750 million at a $44 billion valuation. Four months later, the company is worth $60 billion on paper, a roughly 36% increase in a window most private companies would need years to cover. Axios framed the deal as part of a broader pattern in which a small group of AI-adjacent private companies are being repriced every few months rather than annually, a cadence that was unusual even during the 2021 venture boom.
That speed raises a fair question for anyone who isn’t already a Ramp investor: how much of the jump reflects actual business growth, and how much reflects momentum investing chasing a hot sector. Ramp’s $2 billion annualized revenue figure gives the round at least some grounding in operating numbers, but a $60 billion valuation still implies a revenue multiple that only a handful of software companies have historically sustained.
What Ramp Actually Sells
It’s worth stepping back to what Ramp’s core product actually is, since the AI framing can obscure it. Ramp provides corporate cards and expense-management software, a category that includes competitors built around similar workflows: employees get cards with programmable spending limits, finance teams get automated expense reports, and accounting departments get software that reconciles transactions without manual data entry. That business, not AI, is what generated the $2 billion in annualized revenue Ramp pointed to this week.
Where Ramp has differentiated itself from older expense-software incumbents is in how aggressively it has layered automation, and increasingly AI, on top of that core workflow. The company’s own announcement leaned into that distinction, describing itself less as a card issuer and more as a platform for how businesses spend money, time, and now AI tokens.
“Every Dollar, Hour and Token”: Ramp’s AI Framing
The subhead Ramp used in its own funding announcement, carried by Yahoo Finance, read: “Raises $1.85 billion to help businesses get more value from every dollar, hour and token.” That phrase, specifically the inclusion of “token” alongside dollars and hours, is a deliberate signal. Tokens are the basic unit businesses now pay for when they run AI models, and Ramp is positioning its software as the layer that tracks and optimizes that spending the same way it has long tracked corporate-card spending.
That framing lines up with what Ramp has been building on the product side. The company has described plans to keep developing AI products and to continue hiring, though the exact allocation of the new $1.85 billion toward specific AI initiatives has not been detailed in the public record. What is confirmed is the direction: Ramp wants to be seen as a company that manages AI costs for its customers, not just corporate-card costs.
Inside Ramp Router and the Push Into Developer Tooling
Ramp’s clearest move into coding-tools territory is a product called Ramp Router, described in the company’s own developer documentation as an API for evaluating and routing traffic across different AI models without rebuilding an application each time a team wants to switch providers. Ramp has said internally that using Router cut its own AI costs by roughly 30%, a figure the company has used to pitch Router as something other businesses could adopt for the same reason.
That puts Ramp in a lane adjacent to the AI gateway and model-routing tools that have become a recognizable category this year, the kind of infrastructure that sits between an application and whichever large language model happens to be cheapest or fastest for a given task. It’s a sharply different pitch than plain expense software, and it’s the part of Ramp’s business most directly comparable to the developer-tools companies covered under the coding-assistant market that Claude Code, Codex, and GitHub Copilot currently compete in.
Ramp’s documentation also describes support for connecting AI agents through the Model Context Protocol or a command-line interface, allowing agents to read account data, process approvals, and in some cases make purchases directly. Veeral Patel, Ramp’s head of applied AI, is credited with leading the team behind Router, while Zack Field, a senior staff software engineer, leads what Ramp calls its Agent Developer Platform team, the group building the APIs that let outside AI agents transact on Ramp’s rails.
How Much of Ramp’s Own Code Is Written by AI
The 50-to-80 claim
In a recent podcast appearance recapped online, Ramp’s chief product officer, Geoff Charles, reportedly said that roughly half of Ramp’s own codebase is now AI-generated, with an internal target of pushing that share closer to 80%. If accurate, that would put Ramp among the more aggressive adopters of AI-assisted development at its scale, with product managers increasingly writing specifications for AI agents to implement rather than handing tickets to human engineers first.
Not just Ramp’s problem to solve
That claim fits a pattern documented elsewhere this year. Reporting on how developers now spend their week reviewing AI-generated code rather than writing it from scratch suggests Ramp’s internal numbers, if accurate, track with a broader shift rather than representing an outlier. The harder question, one that applies across the industry and not just to Ramp, is whether a larger share of AI-written code actually produces more working software, or simply more code that still needs human review before it ships.
Agents for Expense Policy and Accounts Payable
Beyond the developer-facing Router product, Ramp has also rolled out autonomous agents aimed at its existing finance customers. According to CNBC’s reporting, Ramp has launched agents that flag unauthorized spending and enforce expense policy automatically, as well as agents for accounts payable that handle invoice coding, approval routing, and fraud detection without a human reviewing every transaction.
Ramp has also described an internal “Voice of the Customer” agent that analyzes customer feedback and assists with the earliest stages of bug triage, a use case that blurs the line between a customer-support tool and an engineering tool. Taken together, the AP agents, the expense-policy agents, and the internal triage agent paint a picture of a company trying to automate both its product and its own operations at the same time, which is precisely the pitch that has drawn comparisons to AI-native engineering organizations rather than traditional fintech incumbents.
Ramp’s $60 Billion Next to Other 2026 AI Mega-Deals
Ramp’s new valuation doesn’t exist in isolation. It lands in the middle of a year in which several AI-adjacent private companies have seen valuations jump sharply within months rather than years. The table below places Ramp’s $60 billion figure next to other large 2026 valuation events reported elsewhere, including some already covered in depth on this site.
| Company | Latest Valuation / Deal | What It Does | Timing |
|---|---|---|---|
| Ramp | $60 billion ($1.85B raised) | Corporate cards, expense software, AI spend tooling | October 2026 |
| Anduril Industries | $90 billion (tender offer) | Defense technology and autonomous systems | 2026 |
| Anthropic | $110 billion annualized run-rate pitch to pre-IPO investors | AI models (Claude) | October 2026 |
| Factory AI | $5 billion | AI coding agents (“droids”) | 2026, tripled in five months |
| Ramp (prior round) | $44 billion ($750M raised) | Same business as above | June 2026 |
What stands out in that comparison is less the absolute dollar figures and more the repricing speed. Ramp moved from $44 billion to $60 billion in a single quarter. Factory AI, a much smaller company by revenue, reportedly tripled its valuation to $5 billion over roughly the same stretch of months, according to earlier reporting on the coding-agent sector. Both data points suggest investors are repricing AI-adjacent software companies on a timeline that looks more like a public-market earnings cycle than a traditional private fundraising calendar.
The AI Coding Tools Boom, in Numbers
Ramp’s bet on AI-generated code and agent tooling is happening against a backdrop of rapid, measurable adoption of AI coding tools across the software industry. Several recent developer surveys quantify just how far that adoption has gone in the past year.
| Metric | Figure | Source |
|---|---|---|
| Developers using or planning to use AI tools in development | 84% | Stack Overflow 2025 Developer Survey |
| Professional developers using AI tools daily | 51% | Stack Overflow 2025 Developer Survey |
| Developers who say debugging AI-generated code is time-consuming | 45% | Stack Overflow 2025 Developer Survey |
| Developers who regularly use AI tools for coding | 85% | JetBrains State of Developer Ecosystem 2025 |
| Developers relying on at least one AI coding assistant, agent, or editor | 62% | JetBrains State of Developer Ecosystem 2025 |
Those numbers help explain why a fintech company like Ramp would build its funding narrative around AI tooling rather than its original card business. If 84% of developers are already using or planning to use AI tools, and roughly half are relying on them daily per Stack Overflow’s survey, then a company that can claim it built its own product with AI, and that it sells software to help other companies manage AI costs, is telling investors a story that matches where the broader developer population already sits. The 45% of developers who told Stack Overflow that debugging AI-generated code is time-consuming is the other half of that story, a reminder that AI-assisted development hasn’t eliminated the need for human review, only shifted where that review happens.
Historical Context: How Fintech Valuations Got Here
Ramp’s valuation trajectory didn’t start this year. The company has been raising money at escalating prices since its early days as a corporate-card startup competing against incumbents like Brex and legacy expense tools built by larger enterprise software vendors. What changed over the past 18 months is the framing: Ramp went from being pitched as a faster, better expense tool to being pitched as an AI-native finance operations platform, a repositioning that tracks almost exactly with when large language models became cheap and reliable enough for companies to build real products on top of them.
That repositioning mirrors what happened at other software categories once AI coding tools matured. Companies that build on top of AI models, or that help other companies manage their AI infrastructure, have generally commanded the valuation multiples once reserved for the model providers themselves. Ramp’s bet is that owning the finance layer of that infrastructure, tracking token spend the way it already tracks card spend, gives it a durable position even if the underlying AI models it routes between keep changing.
What Comes Next: Predictions for Ramp and the Market
A few things seem likely to follow from this round, based on Ramp’s own disclosures and the broader pattern of 2026 AI-adjacent financing.
- Ramp will likely expand Ramp Router and its Agent Developer Platform further, since those are the two product lines most directly tied to the AI narrative that justified this valuation.
- Expect continued hiring, as Ramp has already signaled, particularly on applied-AI and agent-platform teams rather than on the traditional card-operations side of the business.
- Scrutiny of the $60 billion figure will grow if Ramp’s revenue growth doesn’t keep pace with its valuation growth, especially given how short the gap was between the $44 billion and $60 billion marks.
- Competitors in expense management will face pressure to make similar AI and token-spend-tracking claims, whether or not their underlying product changes match the marketing.
- An IPO conversation becomes more plausible the longer Ramp keeps raising at this pace, following a pattern already visible in how Anthropic has pitched investors ahead of its own potential public listing.
None of these are certainties. Ramp has not confirmed IPO plans, and the company’s own statements have been limited to continuing AI product development and hiring. But the shape of this round, a primary-only raise with a long list of public-market-style investors like Fidelity and T. Rowe Price, looks like the kind of financing companies do in the final stretch before going public rather than a typical growth-stage venture round.
Frequently Asked Questions
How much did Ramp raise, and at what valuation?
Ramp raised $1.85 billion in a primary financing round that values the company at $60 billion, announced October 15, 2026.
Who led Ramp’s $60 billion funding round?
Dragoneer Investment Group led the round, with Thrive Capital, Founders Fund, Sands Capital, and Avenir Growth as co-leads, alongside a long list of additional investors including Fidelity Management & Research Company and T. Rowe Price.
How does this compare to Ramp’s previous valuation?
Ramp was valued at $44 billion after a $750 million round in June 2026. The new $60 billion valuation represents a roughly 36% increase in about four months.
What does Ramp actually do?
Ramp provides corporate cards and expense-management software for businesses, along with newer AI products including Ramp Router, an API for routing traffic across AI models, and an Agent Developer Platform for connecting AI agents to Ramp’s systems.
How much of Ramp’s product is built with AI?
Ramp’s chief product officer, Geoff Charles, has reportedly said roughly half of Ramp’s codebase is AI-generated, with an internal goal of reaching 80%, though this figure comes from a podcast appearance rather than an official company filing.
What is Ramp’s annualized revenue?
Ramp said its annualized revenue has topped $2 billion as of the October 2026 announcement.
Has Ramp said it will use the new funding for AI specifically?
Ramp has said generally that it plans to develop more AI products and continue hiring, but has not detailed a specific allocation of the new $1.85 billion toward named AI initiatives.
Is Ramp planning an IPO?
Ramp has not confirmed IPO plans. The structure of this round, heavy with public-market-style investors, has fueled speculation, but nothing has been officially announced.
Related
- Factory AI Hits $5B Valuation, Triples in 5 Months [2026]
- Anduril Tender Offer Hits $90B Value, $1.3B Cap [2026]
- Anthropic Q3 Revenue Hits $16.8B, Jumps 45% Pre-IPO [2026]
- Claude Code vs Codex vs Copilot: 65.5% Dev Share [2026]
- Developers Now Spend 80% of Week Reviewing AI Code [2026]