Product1 publisher3 min readPublished
Instabase renames itself SuperApp and bets the workspace outlasts the extraction pipeline
Instabase, founded in 2015 and funded with more than $250 million, is now SuperApp: team chat, a document library, and automatic model routing. Its document-AI business becomes plumbing.
The Product Desk · Product desk
Drafted by a language model from the sources cited here and checked against its claim ledger before publication. How we use AISend a correction
What happened
- SuperApp Inc., formerly Instabase Inc., announced the launch of its flagship product of the same name and its transition from the Instabase name to the new name.
- The new platform provides teams with a Slack-like communication interface on the web, iOS, Android and Windows, with a document library similar to Google Docs and deep access to AI models from companies including OpenAI Group PBC, Anthropic PBC and Google LLC.
- Users do not need to pick which model to access for efficiency, reasoning, knowledge and cost; the underlying routing chooses the appropriate AI model for the task.
- The company said users can also select a preferred model or compare models in the same context.
- SuperApp is designed around freedom of model choice, data and cloud, with the objective of allowing users to adopt a model without rebuilding applications or moving data, and to avoid architectural or vendor lock-in.
Compiled by The Product DeskSomething wrong?How this is made
Why it matters
Instabase Inc. has renamed itself SuperApp Inc. and shipped a product of the same name: a Slack-like communication interface on web, iOS, Android and Windows, paired with a document library the company compares to Google Docs and access to models from OpenAI, Anthropic and Google [1][2]. The company says this is not a rebrand of Instabase AI Hub but a new product and a new company direction centered on AI collaboration, away from its original vision of unstructured data processing [8].
That last clause is the whole story. Instabase was founded in 2015 and has raised more than $250 million, including a $100 million Series D in early 2025 led by the Qatar Investment Authority [7], which means roughly 40 percent of the company's lifetime funding arrived shortly before it decided the document-processing business was not the thing to build the company around [13]. AI Hub is not being killed; it will continue to receive support and will run on SuperApp technology underneath [9]. The extraction pipeline becomes a dependency of the collaboration surface rather than the other way round.
The mechanical claim is model routing. Users do not choose a model; the platform picks one based on efficiency, reasoning, knowledge and cost [3]. Users who care can still pin a preferred model or compare models side by side in the same context [4]. Around that, the company is selling portability: choice of model, data and cloud, with the stated aim of adopting a new model without rebuilding applications or moving data, and avoiding architectural or vendor lock-in [5]. "The best AI will keep changing. Your work should not have to," founder and Chief Executive Anant Bhardwaj said, adding that "freedom is not a feature, it is the foundation" [6].
The freedom framing is doing double duty. Routing away the model decision genuinely reduces one switching cost, and buyers who signed multi-year commitments to a single vendor have a real problem. But the pitch also asks teams to move their conversation and their documents into a new place, which is a substitution of one dependency for another, not the removal of dependency. The pricing of that trade is not in the announcement.
The product thesis underneath is more interesting than the naming. SuperApp's premise is that people work in one place and talk to AI chats in another, and that this context switching destroys the output of collaboration [10]. Putting the model inside the channel where the team is already arguing is supposed to leave a durable record of research, review, briefing, preview and presentation, including the thinking that led to the answer rather than just the answer [11]. "A private chatbot gives one person an answer," Bhardwaj said. "SuperApp gives the entire team and the world's leading AI models one shared place to turn an idea into finished work" [12].
For operators, the useful reading is a claim about where value settles. A company with a decade of document-AI engineering has concluded that the extraction layer is commodity infrastructure and the record of team work is the defensible product [8][9].