Data Panda lands your business data in one warehouse and lets Cortecs read it. The Vienna-based router exposes one OpenAI-compatible endpoint that fans out to 150+ model endpoints across Mistral, Claude, GPT, Gemini, Qwen, DeepSeek and Kimi, hosted in EU data centres in Spain, Germany, France, Finland and Poland.
Cortecs is run by Cortecs GmbH out of Althanstrasse 4 in Vienna. The product is an AI gateway that sits between your application and the model providers. You call one OpenAI-compatible endpoint, Cortecs filters the providers that match your residency, latency and cost constraints, ranks what remains by price and performance, and dispatches the call. Inference happens in EU data centres in Spain, Germany, France, Finland and Poland, prompts are processed in temporary memory and then deleted, and Cortecs documents that prompts are never stored or used for training.
The catalogue covers more than 150 model endpoints across the names you would expect on a shortlist: Mistral, Anthropic Claude, OpenAI GPT, Google Gemini, Qwen, DeepSeek and Kimi. The same gateway exposes embeddings for search, clustering and recommendation work, image-input handling for the multimodal models, and audio transcription on the providers that ship a speech model. Pricing is pure pass-through. You top up credits, Cortecs takes a flat 5% on the top-up, and the per-token cost on the dashboard is the provider's base rate. There is no monthly subscription, no rate-limit ceiling on the gateway side, and the router publishes a 32% average cost reduction figure based on routing the same workload across providers instead of pinning it to one.
Cortecs spend, model mix and routing decisions per workflow, next to the warehouse content the gateway is reading.
Pipe Cortecs output straight into the systems where the work happens, with the routing layer hidden behind one endpoint.
Cortecs reads what is in the warehouse, the multilingual stack handles the language work, and nothing leaves the EU.
Tools that sit on warehouse data and only call the gateway for the language and document work.
A list of concrete reports, automations and AI features we have built on Cortecs data. Pick the one that matches your situation.
Which model did the gateway pick for each workflow last month, and at what cost?
Routing-decision log per Cortecs call, joined to the workflow that triggered it and to the eventual outcome. Surfaces the workflows where the router keeps switching between providers because none clearly wins on price, and the workflows where it has settled on one vendor for so long that the routing layer is no longer earning its keep. The cost column shows the provider base rate, not a marked-up gateway price.
Did our regulated workloads stay inside the EU region we specified?
Per-call EU-region log, with the workflow's data-classification flag on the same line. Shows which calls landed in Spain, Germany, France, Finland or Poland and which (if any) hit a non-EU fallback the application code did not expect, instead of trusting the dashboard summary at the gateway level.
Is the 32% routing saving real for our specific traffic mix?
Workflow-level cost-per-task rebuild that compares Cortecs' routed cost against the cost of pinning the same workload to a single provider for the same period. Confirms whether the published 32% average shows up in your mix, or whether your traffic is concentrated on the cheapest provider already and the saving is closer to zero.
Cortecs spend per workflow and per business unit, joined to the workflow's measured outcome, with the provider mix the router picked. The AI line on the budget moves from one Cortecs invoice to a number that sits next to tickets handled, briefings shipped and embeddings produced.
Provider-agnostic prompt console for sales enablement: the same briefing or proposal prompt runs on Claude, GPT, Mistral or Gemini through one endpoint, so the rep gets the strongest output for that account without engineering having to re-deploy the application.
Cortecs usage, latency and routing-decision log per workflow on one screen, refreshed daily. The router is followed as a curve, not rediscovered the morning a stakeholder forwards a screenshot of a slow or wrong answer.
These are the 1 tables we currently pull from Cortecs into your warehouse. Query them directly in SQL, join them to the rest of your stack, or build reports on top.
Missing a table you need? We can extend the sync. Tell us what is missing and we will build it for you.
You keep the reporting tool you already have. We connect it to the warehouse where your Cortecs data lives.
OAuth authentication. Read-only by default. We sign a DPA and your admin keeps the keys.
Data flows into your warehouse on your schedule. Near real time or nightly, your call. You own the data.
We build the first dashboard, workflow or AI feature with you, then hand over the keys. Or we stay on for ongoing delivery.
We set up the foundation. Your team builds on top.
Best fit Teams that already have a BI analyst or data engineer and want to own the build.
We build the whole thing, end to end.
Best fit Teams without in-house BI or dev capacity. You tell us what you need and we deliver it.
You do. It lands in your warehouse, on your cloud account. We don't resell or aggregate it. If you stop working with us, the warehouse stays yours and keeps running.
Near real time for most operational systems. For heavier sources we schedule hourly or nightly. You pick based on what the reports need.
No. If you don't have one, we help you pick one and set it up as part of the first delivery. Common starting points are Snowflake, Microsoft Fabric, or a small Postgres start.
Cortecs publishes more than 150 model endpoints across Mistral, Anthropic Claude, OpenAI GPT, Google Gemini, Qwen, DeepSeek and Kimi. The gateway also exposes embeddings, image-input handling for the multimodal models, and audio transcription on the providers that ship a speech model. Every model is reachable through the same OpenAI-compatible endpoint, so swapping a workload from one provider to another is a routing rule rather than a code change.
Cortecs runs in EU data centres across Spain, Germany, France, Finland and Poland. Cortecs documents that prompts are processed in temporary memory and deleted immediately, and are never stored or used for training. The product is positioned as a GDPR-compliant router with EU-only hosting, which is why regulated buyers in financial services, public sector and healthcare end up shortlisting it when residency is a hard constraint.
There is no subscription. You top up credits, Cortecs takes a flat 5% on the top-up, and the per-token price on the dashboard is the provider's base rate with no markup. Cortecs publishes a 32% average cost reduction figure based on routing the same workload across providers instead of pinning it to one. Whether that average lands for you depends on your traffic mix, which is why we wire the routed cost back into the warehouse and rebuild it at workflow level.
We review your Cortecs setup and the systems around it. Together we pick the first thing worth building.