Skip to content
World clockEU--:--UK--:--USA--:--CN--:--PLDEFRIT中文EN

portal about AI and technologyevents · analysis · interviews · technical background

Search
LIVE
›

Wabi raises $55M at a $1.25B valuation for its personal AI agent

Wabi, the startup founded by Replika creator Eugenia Kuyda, closed a $55 million round at a $1.25 billion valuation on 29 October. IVP led the round, with Andreessen Horowitz, Forerunner and the Amazon Alexa Fund taking part.

AI & modelsExplainerRachel NwosuPublished: 5 November 20263 min readSources 17
Wabi raises $55M at a $1.25B valuation for its personal AI agent

Wabi is now valued at $1.25 billion, above the $1 billion mark that defines a unicorn, Whales Book reported. The bet rests on Kuyda's record and the product rather than on revenue, Crypto Briefing reported.

The round

IVP led the $55 million round, with Andreessen Horowitz, Forerunner and the Amazon Alexa Fund participating. The angel list includes Shopify's Tobi Lütke, Salesforce's Marc Benioff, former Yahoo head Marissa Mayer, former Twitter chief executive Dick Costolo and Alex Pall of The Chainsmokers, AI Insider reported. The Amazon Alexa Fund is the investment arm attached to Amazon's voice assistant business, Crypto Briefing noted. The round closed on 29 October 2026. It follows a $20 million pre-seed in November 2025 that Andreessen Horowitz led, taking Wabi's total funding to $75 million in about 11 months. Andreessen Horowitz stayed in as a participant after IVP took the top slot.

Alex Lim, a general partner at IVP, said Kuyda has a distinctive sense of where AI is heading and a strong track record of building it. He said her view of how people will interact with AI goes well beyond the text box or command line, AI Insider reported.

The pivot

Wabi did not start as an agent company. Its original product let anyone build apps by writing prompts. Wabi 2.0, announced in late September 2026, replaced that with a messenger-style agent that builds the interfaces a user needs to get a task done. It supports multi-thread chats, shared access between users and mini apps created on demand, Crypto Briefing reported. The company says the agent can build tools for meal planning, tracking local restaurants, planning date nights, designing fitness plans and custom trivia games, AI Insider reported. The full public launch came in mid-to-late October 2026, so the round closed only days or weeks after ordinary users could try the product.

The shift reflects a broader industry move toward models that execute multi-step workflows, Whales Book said.

The founder

Kuyda previously built Replika, the AI companion app that has drawn roughly 35 to 40 million users, Crypto Briefing reported. She stepped down as Replika's chief executive in early 2025 to start Wabi. Investors are pricing Wabi on its product, its founder and its trajectory, not on revenue.

A crowded field

Wabi is not the only agent company pulling in money. Instinct raised $1 billion at a $10 billion valuation, and Tolan closed a $30 million round for its AI companion, AI Insider reported. Whales Book said the sector is becoming crowded, with large global platforms and well-funded startups racing to build systems that can handle personal tasks. Wabi will have to compete with established platforms for user adoption.

The open question

Wabi remains free and has no finalized monetization model, Crypto Briefing reported. Whales Book said the valuation will depend on whether Wabi can move past the prototype stage to deliver a reliable, functional interface that gives users tangible value. Pivoting a business model after building the first product is risky, because it requires shifting both the technology and user behaviour, the outlet said.

The bet is that people want an assistant that assembles its own interfaces rather than answering in a text box. Whether they will pay for one is a question the funding did not answer.

Comments 0

Sources

17
  1. 01Wabi Secures $55M at $1.25B Valuation After Pivot to Personal AI AgentEN
  2. 02AI Startup Wabi Hits $1.25 Billion Valuation After PivotEN
  3. 03Replika founder's Wabi raises $55M at $1.25B valuation as it pivots to AI agentsEN
  4. 04AI Startup Wabi: సరికొత్త వ్యూహంతో 'యూనికార్న్' క్లబ్‌లోకి.. $1.25 బిలియన్లTE
  5. 05Elon Musk Tightens Grok Bot Usage Limits, Citing Unsustainable Costs - AI InsiderEN
  6. 06Zipline raises $1.5bn at $21.5bn valuation as Walmart adds 300 drone sitesEN
  7. 07Baidu releases ERNIE 5.5 Flash Base with 280 billion parameters as an open-source foundation modelEN
  8. 08Shield AI buys radar maker Echodyne for $5.3B at a $28.8B valuationEN
  9. 09Zipline hits $21.5B valuation as Walmart expands bet on drone deliveryEN
  10. 10Elon Musk’s nearly half a billion dollars in midterms spending was a flopEN
  11. 11Fal doubles valuation to $16B after $1.3B Series FEN
  12. 12Revolut confirms $115BN valuation, after completing secondary saleEN
  13. 13China blocks SoftBank’s US$5.4 billion takeover of ABB’s robotics divisionEN
  14. 14$150 Billion More Paid by Americans for Gas This Year Than LastEN
  15. 15Binance Reveals How Fake Job Interviews and AI Tool Abuse Led to $6.43 Billion TheftEN
  16. 16FBI Links North Korea's TraderTraitor to Record $6.43 Billion Binance Crypto HeistEN
  17. 17ByteDance’s Zhang Yiming tops China rich list with US$99.7 billion as AI fuels fortunesEN

All figures and quotations in this text come from the sources listed below.

Content prepared by the editorial team with AI assistance.

Rachel Nwosu

Rachel Nwosu

AI, models and technology

Rachel Nwosu covers AI, models and technology for FLASH24, working from public model documentation, benchmark releases and repository histories rather than press summaries, and she skips announcements that arrive without reproducible numbers. She checks training-data claims against dataset cards and reruns reported metrics where code is available. She spends much of her week interviewing researchers and engineers, tracking model launch calendars, and comparing vendor benchmarks with independent evaluations. Outside the desk she runs 3D printers, restores old computers, and tests how models learn from internet junk. She does not publish benchmark figures she cannot trace to a source.

Newsroom →

Comments

0
  1. No comments yet — be the first.

Write a comment

Comments are public. We do not publish abuse, spam or advertising.