Skip to content
AgentsIndex

TextCortex

Berlin-built no-code agent platform with EU hosting, a choice of models across vendors, and an on-premise option.

TextCortex is a Berlin-based enterprise platform for building AI agents without code, grounding them in company knowledge and scheduling them as automations. Agents run on a choice of third-party models, can be called from Slack or Teams, and the platform can run on-premise.

Best for
Enterprise teams who need AI agents tied to company knowledge and workflows.
Look elsewhere if
  • you want to use it for real without paying
  • you want any plan without talking to sales
  • you need single sign-on below the Enterprise plan

What we checked

Pages last read
Facts checked
17 facts

From 6 sources

AI answers
Too few read

58 buyer-question answers read, November 2026

At a glance

Tags
Free trial, GDPR, ISO 27001, Self-hosted, SOC 2

For developers

Works with
Slack, Microsoft Teams, SharePoint, Outlook, Microsoft Suite, Webflow and 24 more

What it actually is

TextCortex ships two things: a browser extension/desktop app that injects AI writing, rewriting and chat into any text field across roughly 30,000+ sites (its own count), and a web platform for building no-code "agents" that can be scheduled, @mentioned in Slack/Teams, and pointed at connected knowledge sources (docs, wikis, Salesforce, HubSpot, etc.). It is model-agnostic: the same account can call Claude, GPT, Gemini, Qwen, DeepSeek, Kimi, or GLM depending on task and budget, rather than being tied to one vendor's model. (https://textcortex.com/, https://textcortex.com/integrations)

Pricing and how the usage cap works

Free is $0/month: knowledge bases, custom agents, deep research, and 50MB storage, with 100 creations on signup and 20/day after (1 "creation" ≈ 125 words). Premium is $29.99/user/month (excl. VAT), billed either as a flat rate with a credits slider (150 to 2,800 creations/month) or Unlimited. Enterprise is quote-only, adding SSO/SCIM/SAML, dedicated deployment and centralized billing. A 14-day trial exists but there is no seat-sharing on individual accounts. (https://textcortex.com/pricing)

The more interesting design choice: every seat's Premium price funds a token wallet worth 50% of the license fee for premium-model usage (frontier models like GPT-5.5 or Claude Opus, deep research, long agent runs). Standard models keep working after the wallet is empty — no hard lockout, no surprise overage bill unless an admin explicitly turns on paid "Extra Usage." TextCortex publishes the underlying per-model token costs (e.g. Claude 4.6 Sonnet at €2.80 (about $3.08)/€14.01 (about $15) per 1M input/output tokens, GPT-5.5 at €4.67 (about $5.13)/€28.01 (about $31)) rather than hiding them behind a vague "credits" abstraction. (https://help.textcortex.com/hc/en-us/articles/46798075584657-Model-usage-limits-and-Extra-Usage)

Data handling and EU hosting

TextCortex hosts services in the EU (GDPR-compliant infrastructure, with non-EU model calls covered by SCCs or the US-EU Privacy Framework) and states customer data is never used to train or fine-tune models. It maintains a public trust center at trust.textcortex.com and lists SOC2/ISO certification on its enterprise page, though the certificates themselves are not independently viewable from that page. Where the help center publishes model-hosting regions, several premium models (Claude, Qwen, GLM) do route through EU or UK data centers; a few (GPT-5.5 family via NL, DeepSeek via GB, some Gemini/Kimi variants via US) do not. (https://textcortex.com/business/enterprise, https://help.textcortex.com/hc/en-us/articles/46798075584657-Model-usage-limits-and-Extra-Usage, https://trust.textcortex.com/home)

Company size and the funding gap

TextCortex was founded in January 2021 by Ceyhun Derinbogaz and Dominik Lambersy. The only publicly disclosed funding round is a $1.2M pre-seed in June 2022, led by btov Partners (b2venture) with participation from Speedinvest, Entrepreneur First, and angels including Wayve co-founder Amar Shah. No later or larger round has been announced in the press coverage available. (https://www.startbase.de/news/textcortex-erhaelt-12-millionen-us-dollar-pre-seed-finanzierung/, https://www.b2venture.vc/stories/turbocharging-written-communication-with-artificial-intelligence)

The current marketing — "2.4M users worldwide," "$550K saved for every 100 enterprise users/year" from a self-described "study of 2,000 enterprise users," 75% enterprise adoption within 3 months — are all vendor-stated figures on the homepage with no named methodology or independent citation. A buyer sizing TextCortex as a long-term enterprise infrastructure vendor should treat the funding level and the scale claims as two facts that don't obviously reconcile, and ask for references directly rather than relying on the homepage stats. (https://textcortex.com/)

Independent reviews

TextCortex is rated 3.7/5 ("Average") on Trustpilot. Reviews split by era: 2023–2024 reviews are largely positive about the browser-extension writing tool (tone-changing, multilingual support, ease of use). Reviews from 2026 are more negative and enterprise/billing-focused: one from August 2026 describes 18 repeated card-charge attempts after cancellation with unresponsive support; another from July 2026 reports previously-available models (Claude Sonnet 4.6) being restricted without notice and a two-week support wait. (https://www.trustpilot.com/review/textcortex.com)

TextCortex's own pricing-page schema markup claims a 4.8/5 rating from 257 reviews and cites G2 recognitions ("Top 50 AI Products 2025," "Top 100 Highest Satisfaction Products 2025"); we could not confirm these figures on G2 or Capterra, so they are listed here as vendor-stated only.

Where it fits

If the use case is an individual or small team wanting a model-agnostic writing/rewriting assistant that works inline across the web via browser extension, TextCortex's free tier and $29.99/seat Premium are competitively specified against single-model competitors, and the usage-wallet design avoids nasty overage surprises. If the use case is deploying "enterprise AI infrastructure" at scale — the positioning TextCortex now leads with — the funding history, unverifiable scale claims, and the 2026 billing/support complaints are reasons to run a paid pilot and check references before committing to Enterprise, rather than treating the case studies (MAHLE, b2venture) as representative.

TextCortex price and terms

Terms

Free trial
Yes14 days
Sign up without sales
PartlySome plansUp to Premium

For AI agents

Developer docs

What needs a higher plan

Enterprise
Single sign-on
TextCortex's pricing page

Pricing page, captured

web.archive.org/web/20261111085022/https://textcortex.com/pricing, checked

What we checked on TextCortex

YesPartlyNo

What it costs to startYes

What the price counts

Priced per seat: Premium is a per-user monthly price that includes some AI model usage, while Enterprise is sold on a custom quote with dedicated deployment.

$29.99$23.99/ user/month Billed monthlyBilled annually

web.archive.org/web/20261111085022/http…Checked

What it costs to startPartly

Free to use

The current pricing page lists only Premium and Enterprise, and the free way in is a 14-day trial rather than a standing free plan.

Yes. TextCortex offers a 14-day free trial so you can explore its features.

web.archive.org/web/20261111085022/http…Checked

What it costs to startPartly

Self-serve signup

Premium can be bought online at the published per-user fee without a sales call, but Enterprise features such as SSO and dedicated deployment go through a demo request.

By signing up for a paid TextCortex plan, you agree to pay TextCortex the monthly or annual subscription fees published on the Website as of when you…

web.archive.org/web/20261008094338/http…Checked

What it costs to startYes

Contract

Premium is sold on either a monthly or an annual subscription that renews automatically, so month-to-month billing is available without a yearly commitment.

Your monthly or annual subscription will automatically renew unless you cancel it.

web.archive.org/web/20261008094338/http…Checked

What it costs to startYes

Cancel any time

Cancellation is self-serve from the billing dashboard at any time and takes effect at the end of the paid period; upgrade fees are non-refundable beyond a 5-day guarantee on certain purchases.

You may delete your account or cancel your subscription at any time through the Settings section of your Billing dashboard.

web.archive.org/web/20261008094338/http…Checked

What it works withYes

Works with

The integrations page names connectors including Salesforce, HubSpot, Microsoft Teams, Google Drive, OneDrive, Jira and Asana, plus Zapier and Make; the plan for each is not stated.

Use TextCortex directly on over 30.000+ platforms or connect AI with 5000+ more apps via Zapier or Make.

web.archive.org/web/20260829023139/http…Checked

What it works withYes

Public API

The pricing FAQ confirms a text generation API with self-serve keys from account settings; which plans include it is not stated on the page.

Yes, we have a Text Generation API. You can generate an API key in your account settings and get started right away.

web.archive.org/web/20261111085022/http…Checked

What it does on its ownYes

Acts on its own

Agents can be scheduled to run recurring work on their own while the user is offline, delivering results by email or in-app notification.

Schedule agents for recurring work Agents work even when you're offline Get results to your inbox daily/weekly/monthly

web.archive.org/web/20261113163339/http…Checked

What it does on its ownYes

Limits on what it may do

IT controls who has access, which models are allowed and where a human must sign off; the page does not say which plan carries these controls.

IT sees who has access, which models are allowed and where human signoff is required.

web.archive.org/web/20261113163339/http…Checked

What it does on its ownYes

No code needed

Agents are configured in a visual builder with model, tools and knowledge picked from menus, and automations are built in a node-based visual editor.

Create and deploy intelligent agents through a simple visual interface, without writing a single line of code.

web.archive.org/web/20261113163339/http…Checked

What happens to your dataYes

No training on your data

The vendor states that customer data is never used to train or fine-tune models, and repeats 'zero data training' on its agents page.

Absolutely not. None of your data will ever be used for model training or fine-tuning.

web.archive.org/web/20261111085022/http…Checked

What happens to your dataPartly

Data stays in your region

The shared service is hosted only in the EU, with no region choice, while the separate On-Premise line runs in the customer's own data center.

We host our services exclusively in the European Union on GDPR-compliant servers, including third-party AI models.

web.archive.org/web/20261111085022/http…Checked

What happens to your dataPartly

Self-hosted

A separate TextCortex On-Premise offering runs on GPU servers in the customer's data center with open-weight models, priced by quotation rather than on the standard plans.

Run TextCortex on servers in your own data centre. Set your headcount, usage and open-weight model, and see the GPU servers, load and electricity…

web.archive.org/web/20261008094132/http…Checked

Whether a company can buy itYes

Security audits

The trust center names completed SOC 2 Type I, SOC 2 Type II and ISO 27001 audits, with reports available on request.

Text Cortex AI has successfully completed a Service Organization Control 2 (SOC 2) Type II audit of our control environment via an independent…

trust.textcortex.com/homeChecked

Whether a company can buy itPartly

Single sign-on

SSO, SCIM and SAML are listed only on the Enterprise plan, which is sold on a custom quote; Premium does not list them.

Everything included in Premium SSO / SCIM / SAML Dedicated deployment Centralized billing Priority support

web.archive.org/web/20261111085022/http…Checked

Models it can useYes

You pick the model

Users pick the model for each agent from a hub spanning Claude, GPT, Gemini and DeepSeek families, and the vendor promotes no lock-in to one provider.

Secure access to all major models. No lock-in with one model provider.

web.archive.org/web/20261111085022/http…Checked

Models it can useYes

Open models

The On-Premise line runs open-weight models such as DeepSeek, GLM, Qwen and Gemma, and DeepSeek V4 also appears in the hosted model hub.

Compare DeepSeek, GLM, MiMo, Kimi, Gemma, Qwen and Nemotron on quality, memory and speed before you buy hardware.

web.archive.org/web/20261008094132/http…Checked

TextCortex alternatives

Questions about TextCortex

How much does TextCortex cost?
TextCortex costs $29.99 a seat a month. From web.archive.org/web/20261111085022/https://textcortex.com/pricing, checked 17 November 2026.
Does TextCortex have an API or an MCP server?
TextCortex has an API. From web.archive.org/web/20261111085022/https://textcortex.com/pricing, checked 17 November 2026.
Is TextCortex open source, and can you host it yourself?
You can host part of TextCortex yourself. From web.archive.org/web/20261008094132/https://textcortex.com/on-premise, checked 17 November 2026.
What have we checked on TextCortex?
17 facts, under these headings: What it costs to start, what it works with, what it does on its own, what happens to your data, whether a company can buy it, models it can use.
How do we check TextCortex?
We hold 17 facts on TextCortex, quoted from 6 pages. Facts and price last checked 17 November 2026, when we also found the last change.

How we check every listing

TextCortex price history

We have read the price once so far, on 17 November 2026: from $29.99 a seat a month.

PricePlanFirst seenLast seenReadingsSource
$29.99 a seat a month17 November 202617 November 20261Pricing page, read 17 November 2026

From our research on 17 November 2026. A date is when we read the page, not when the price changed.

Similar in Workflow automation

From our research

All research

Is TextCortex yours?

Free, and the page then carries the verified owner mark. You confirm it with an email at textcortex.com. Claim this listing

Report an error if a fact is wrong.