Choose Your AI Provider
OpenEduCat does not bundle an AI provider. Instead, the Bring Your Own Model (BYOM) feature lets you connect whichever provider your institution has already approved, budgeted for, or is required to use under your data governance policy.
Every major provider is supported out of the box. Pick the one that fits your compliance requirements, budget, and technical setup, then connect it in minutes. All 9 AI tools in OpenEduCat will automatically use your chosen provider.
Supported Providers
Seven providers supported natively. Plus any server that exposes an OpenAI-compatible endpoint.
OpenAI (GPT-4o)
Most popularMost widely used API-first provider. Strong general-purpose performance across all 9 AI tools.
See integration guide →Google Gemini
MultilingualMultilingual support and large context windows. Integrates naturally with Google Workspace environments.
See integration guide →Microsoft Azure OpenAI
EnterpriseEnterprise compliance path. FERPA-friendly infrastructure, Azure AD integration, full audit logs.
See integration guide →Anthropic Claude
Long contextLong-context document analysis and safety-focused reasoning. Ideal for IEP drafting and nuanced feedback.
See integration guide →Meta Llama
Self-hostedOpen-weight models for self-hosted deployments. Zero data egress, all processing stays on your servers.
See integration guide →Mistral AI
EU data residencyLightweight, efficient, and EU-hosted. GDPR-native with European data residency for privacy-first institutions.
See integration guide →AWS Bedrock
Multi-modelMulti-model access via a single AWS endpoint. Ideal for institutions already running on AWS infrastructure.
See integration guide →How provider switching works
Changing your AI provider takes under two minutes. No code changes, no migration, no downtime.
Get your API key
Obtain an API key from your chosen provider. OpenAI, Anthropic, Google, and Mistral all offer self-service API key generation. For Azure and AWS, configure through your existing cloud console.
Paste it into OpenEduCat
In the OpenEduCat admin panel, navigate to AI Settings > Provider Configuration. Select your provider, paste your API key, and optionally specify a preferred model version. Save.
All 9 tools use your provider
Immediately, every AI tool in OpenEduCat routes through your chosen provider, the grading assistant, worksheet generator, IEP writer, course builder, and the rest. Switch providers again anytime.
Which provider is right for your institution?
The decision usually comes down to three factors: data compliance requirements, existing vendor relationships, and cost.
If data residency or sovereignty is the primary concern
Use Meta Llama (self-hosted on your own servers, zero data leaves your network), Mistral (EU data centers, GDPR-native), or Microsoft Azure OpenAI (in-region processing with enterprise compliance agreements).
If you already have Microsoft or AWS contracts
Azure OpenAI Service and AWS Bedrock let you route AI traffic through your existing enterprise agreements, BAAs, and IAM policies. No new vendor relationships to establish.
If performance across the widest range of tasks is the goal
OpenAI GPT-4o consistently performs well across all 9 AI tools, essay grading, quiz generation, lesson planning, IEP drafting, and student support. It is the most widely tested provider against educational workflows.
If multilingual or large-document processing matters
Google Gemini handles multilingual prompts and large context windows well. Anthropic Claude is purpose-built for long-document reasoning, important for IEP processing and policy document analysis.
If cost efficiency on a tight budget is the priority
Mistral offers efficient, lower-cost inference with European hosting. For institutions with GPU resources, self-hosted Meta Llama has no per-token cost after the hardware investment.
Ready to connect your AI provider?
Book a demo and we will walk through provider configuration live, including how to set rate limits, monitor token usage by department, and switch providers without disruption.