Innogath
Innogath is an AI research workspace that turns a question into a cited, multi-chapter report, an auto-generated diagram and a tree of branching pages, each inheriting its parent's research context, exportable to PDF, Markdown or DOCX.
What is Innogath?
Innogath is a research workspace built on a premise its homepage states outright: AI conversations are linear, while human thought branches. Instead of a chat log, the product organises work as a tree of pages, each one a complete research artefact.
The core action is Deep Research. From one prompt the system plans the work, reads the evidence (open web pages, uploaded PDFs, CSVs, spreadsheets and images, or earlier reports in the same project), cross-checks claims, and writes a report of numbered chapters with inline citations and a consolidated references section. The publisher says a run reads roughly 20 to 50 sources and takes several minutes. An Activity Stream shows planning, searching, source-reading and drafting as they happen, which doubles as an audit trail of where an answer came from.
Every report is branchable. Select a passage, click Reply, and a child page appears under that section with the quoted text as context; the child inherits a summary of every ancestor's research, so follow-ups build on established facts instead of restarting. A Pages Overview tree and breadcrumbs keep the whole project navigable.
Each run also produces a knowledge diagram on a canvas beside the report, drawn from one of 22 visual formats, among them flowchart, mindmap, SWOT, timeline, matrix, Venn and funnel. The diagram can be regenerated, edited in an Excalidraw-style editor and exported as a high-resolution PNG, and clicking any node opens a focused chat or spawns a new branch.
A project Notebook closes the loop: a rich-text editor with slash commands, task lists, code blocks and embedded video, fed by a Send to Note action that carries the source reference and a backlink. Reports export to PDF, Markdown and DOCX with citations intact.
Four chat modes, Auto, Fast, Thinking and Deep Research, let you trade speed against depth, and usage runs on monthly credits rather than message counts. The product works in the browser and installs as a desktop application on macOS and Windows through a progressive web app, with an interface available in nine languages. It is young: the public changelog runs from version 0.8.0 in February 2026 to version 0.18.0 in May 2026.
What it does
- Turn one question into a multi-chapter report with inline citations, drawn from 20 to 50 sources
- Branch any sentence of a report into a child page that inherits the parent's research context
- Generate an editable knowledge diagram from the report, chosen among 22 visual formats
- Ground answers in your own PDFs, CSVs, spreadsheets, images and links
- Collect findings in a project notebook, each passage carrying its source reference and a backlink
- Export reports to PDF, Markdown or DOCX with citations preserved, and the canvas as a PNG
- Navigate an entire project as a tree through the Pages Overview and breadcrumbs
When to use Innogath / When not to
A quick filter to help you decide if Innogath is the right fit.
When to use Innogath
- PhD candidates and postdoctoral researchers running literature reviews that have to survive supervision, revision and challenge
- Strategy consultants and corporate-development analysts assembling market scans and client briefs from dozens of sources
- Investigative and data journalists who need a traceable source trail behind every claim they publish
- Product and competitive-intelligence managers tracking competitors, pricing and regulation as separate, revisitable branches
- Students writing source-grounded essays and dissertations whose citations must still check out months later
When not to use Innogath
- Teams that need an identifiable contractual counterparty: no legal entity, registered address or governing jurisdiction is published anywhere on the site
- Organisations bound by data-residency or certification requirements, since no hosting country is named and no SOC 2 or ISO 27001 audit is claimed
- Developers who want to automate research or plug it into an existing pipeline: there is no public API and no documented integration surface
- Mobile-first users, as there is no iOS or Android application, only a web app and a progressive web app for macOS and Windows
- Anyone after quick conversational answers, a case the publisher's own comparison page hands to a general-purpose chat assistant
How to use Innogath
A typical end-to-end flow, from setup to results.
- Create an account at app.innogath.com with an email address or a Google account, then click the verification link
- Follow the eight-step guided tour of the workspace, replayable later from the settings panel
- Create a project from the sidebar with New Project
- Type your research question in the chat box, as specifically as you can phrase it
- Attach PDFs, CSVs, documents, images or links from the plus menu to ground the answer in your own material
- Press Enter to let Auto mode choose the strategy, or click Deep Research to force a full report
- Follow the Activity Stream while the system plans, searches, reads sources and drafts
- Select any passage of the report and click Reply to branch a sub-question into its own page
- Click a node on the canvas to open a focused chat, refresh the diagram, or edit it by hand
- Send passages to the Notebook, write your deliverable there, then export to PDF, Markdown or DOCX
Pros & Cons
Pros
- Every claim carries a clickable citation, and evidence drawn from uploaded files opens in a dedicated modal
- Branching with inherited context keeps a long project navigable where a chat thread would bury it
- Report, diagram and notebook sit on the same research tree instead of in three separate tools
- Sub-processors are named one by one in the privacy policy: Supabase, Google, OpenAI, Anthropic, Moonshot AI, DeepSeek, Serper and DODO Payments
- A clear position against training on private workspace content, backed by provider-side configuration
- Permanent free plan of 500 credits a month with no card required, plus self-service data export and account deletion
- Dated public changelog with monthly releases, and a comparison page that concedes when a rival is the better fit
Cons
- No legal entity, registered address or governing jurisdiction is published; the terms defer to an unnamed contracting entity
- No hosting country or region is disclosed, and no security certification such as SOC 2 or ISO 27001 is claimed
- No EU representative under Article 27 and no data protection officer are designated
- No public API, no mobile application and no browser extension; the desktop app is a progressive web app for macOS and Windows only
- A single address, support@innogath.com, handles support, legal and privacy requests; there is no contact page and no help centre
- Advertised usage figures are undated, unsourced and mutually inconsistent, on a domain registered in February 2026
- Headline prices assume annual commitment, and four of the six advertised use-case scenarios are still marked as forthcoming
Pricing & Plans
Innogath offers a permanent free plan with 500 credits per month and no payment card required. The lowest paid entry point is the Pro plan at USD 9.60 per month, billed annually, with the Ultra plan at USD 32 per month on the same terms. Prices are stated exclusive of applicable taxes, subscriptions renew automatically until cancelled, and a new paid subscription may be refunded in full within 14 calendar days of the initial charge.
- 500 credits monthly
- basic research depth
- Fast and Auto chat modes
- auto-generated diagrams
- notebook and canvas
- Markdown export
- one concurrent research
- 5
- 000 credits monthly
- everything in Free plus standard research depth
- all four chat modes including Thinking and Deep
- PDF and DOCX export
- diagram refresh
- auto-citations and source linking
- three concurrent researches
- 20
- 000 credits monthly
- everything in Pro plus maximum research depth
- fastest processing speed
- unlimited concurrent researches
- priority support
Data, GDPR & hosting
A consolidated view of how Innogath handles your data.
GDPR overview
The GDPR is addressed concretely rather than name-checked. The privacy policy, in force since 9 April 2026, identifies Innogath as data controller, sets out five legal bases for processing (contract, legitimate interests, legal obligation, consent, vital or public interest) and eight data-subject rights, including erasure, portability and complaint to a supervisory authority. Rights are exercised by email to support@innogath.com. Transfers out of the EEA, the United Kingdom and Switzerland rely on the European Commission's Standard Contractual Clauses, the UK IDTA or its addendum, with copies available on request. A Data Processing Agreement is offered to organisational customers, and sub-processors are listed by name. Two gaps deserve mention: no EU representative is designated under Article 27, and no data protection officer is named. The site never claims compliance in so many words.
Who owns the data?
Under the Terms of Service you keep all ownership rights in your User Content, meaning prompts, files, notes, diagrams and canvases, and as between you and Innogath you also own the AI Output generated from them. Innogath takes a worldwide, non-exclusive, royalty-free, sublicensable licence limited to four stated purposes: running the service, security, abuse prevention and debugging, legal compliance, and product improvement in aggregated or anonymised form. Two reservations matter. Feedback you submit is assigned to Innogath under a perpetual, irrevocable licence, without compensation or attribution. And identical prompts may produce identical output for other users, so no exclusivity is guaranteed.
Reuse rights
You may reuse your reports and AI Output freely, without asking Innogath for permission. The Terms place ownership of AI Output with you and make you responsible for how it is used downstream, including publication, distribution and commercial exploitation, so a report can become a client brief, a thesis chapter, an article or a deck. Two limits apply. The publisher gives no warranty of accuracy or non-infringement and states plainly that AI output can be incomplete, inaccurate, outdated or biased, so independent verification is required before any legal, financial, medical, regulatory, academic or operational decision. Output is also not guaranteed to be unique to you. Separately, prompts and the User Content you choose to include are transmitted to third-party model providers, namely OpenAI, Anthropic, Moonshot AI and DeepSeek, acting as processors under contract; Innogath states it does not use private workspace content to train public foundation models.
Data retention & training
Hosting summary
No hosting country or region is named anywhere on the site, neither in the privacy policy nor in the terms, and there is no trust or security page. That absence is the main fact to record. What the publisher does disclose is its supply chain: Supabase provides the database, authentication and object storage, Google handles federated sign-in, OpenAI, Anthropic, Moonshot AI and DeepSeek run inference, Serper serves the web-search API, and DODO Payments acts as merchant of record. Cloud infrastructure and content delivery appear as categories, without a named vendor. Innogath states that it operates globally and that personal information may be transferred to and processed in countries other than the user's own, relying on the European Commission's Standard Contractual Clauses, the UK International Data Transfer Agreement or its addendum, and supplementary measures where required; copies of those safeguards are available on request. Encryption in transit is described, encryption at rest is not mentioned. An organisation with a data-residency requirement therefore has nothing to rely on beyond that transfer framework, and should raise the question before committing.
Things to keep in mind
Risks and trade-offs to weigh before adopting Innogath.
- The publisher operates behind a trade name only: with no registered entity, no address and no named jurisdiction, there is no identifiable counterparty should a dispute arise
- Liability is capped at the greater of twelve months of fees or USD 100, and both the service and its AI output are supplied as is, with no warranty of accuracy or availability
- Citations make output look verified; they show where a claim came from, not that it was read correctly, and the documentation itself asks that important decisions be checked against the underlying source
- A tool that returns a finished-looking report in minutes invites outsourcing the judgement research is meant to build, and the branching tree can make navigational fluency feel like understanding
- Features, plans, quotas and the underlying models may change without notice, accounts may be suspended after prolonged inactivity, and users are solely responsible for their own backups
- Deleted content persists in backups for a limited further period, and authorised staff may review specific AI interactions when abuse is reported
- The supply chain is long, with eight named third parties across storage, authentication, inference, search and payments, and the marketing site carries a Meta tracking pixel
Setup & Integrations
Technical difficulty
Very low. Innogath runs in the browser at app.innogath.com: sign up with an email address or a Google account, click the verification link, and an eight-step guided tour opens on first use. A first project takes three actions: create it from the sidebar, type the question, press Enter or click Deep Research. No API key to obtain, no model to configure, no data source to connect. The desktop app installs as a progressive web app from the browser, around 2 MB with no installer file and automatic updates, requiring macOS 12 or Windows 10 or later.
Deployment
Integrations
Supported languages
Behind Innogath
Resources
All the official URLs gathered for verification and reference.
Alternatives
Tools that compete with or complement Innogath.
Frequently asked questions
What is a credit, and how far does one go?
Can I try Innogath before paying?
What happens when I run out of credits?
Are the sources really verifiable?
Is my content used to train AI models?
In which formats can I export my work?
Is there a mobile app or an API?
Where is my data hosted, and what is the minimum age?
Should you pick Innogath?
Innogath answers a real complaint about AI research: the chat thread, where the twentieth question buries the second and nothing survives the session. Its response is structural rather than cosmetic. A question produces a cited report; any passage of that report branches into a child page that already knows what the parent established; a diagram is drawn from the result; a notebook turns the whole tree into a deliverable. For work that has to be defended after the first read, that shape is genuinely useful, and the citation trail plus the Activity Stream make it auditable rather than merely plausible.
The product side holds up. Nine interface languages, three export formats with citations intact, 22 diagram types, cross-device resumption of a running search, two-factor authentication, self-service data export and deletion, and a dated changelog shipping monthly. The privacy policy is unusually forthcoming for a young tool, naming eight sub-processors individually and taking a clear position against training on private workspace content.
The reservations are about the company, not the software. No registered entity, no postal address, no named jurisdiction, no hosting country, no security certification, no EU Article 27 representative, and a single email address covering support, legal and privacy. The terms defer to a contracting entity they never identify. Headline usage numbers are undated and inconsistent with the site's own monthly counter, on a domain registered in February 2026 and a product still at version 0.18.
For an individual researcher, consultant, journalist or student on a free or nine-dollar plan, that asymmetry is acceptable and the workspace is worth trying. For an organisation that needs a contract, a data-residency guarantee or an audited supplier, it is not yet a candidate. Try it on a real topic, and judge the branching before the marketing.
- Choosing a selection results in a full page refresh.
- Opens in a new window.