The fastest way to build Retrieval Augmented Generation (RAG) applications. Compliant. Controlled. Connected
Credal can be deployed into your environment, or we can host it in ours. If we host it, you can get a single tennant instance or multi-tennant instance.
Developers need to be able to access the core functions of their vector store - searching, inserting, querying and deleting data, without unneccessary abstractions getting in the way. Credal combines the flexibility of direct, low level APIs, with the simplicity of carefully designed abstractions.
Credal’s APIs support Keyword, Semantic and Hybrid search out of the box.
All requests to Credal’s vector store are audit logged, alongside the user making the request, and the returned results. Our Audit Logs are API exportable for integration with your SIEM and your monitoring tools
One of the hardest parts of securing a vector store is a good way to enforce permissions on your vector database. Credal can automatically enforce the end-user permissions when making queries, via either OAuth, or an Admin approved service user that can simply pass in the user email.
Built-in compatibility with GDPR compliance workflows like Data Deletion, and automated Data lineage tracking.
No Code data connections for key sources like GDrive, Slack, Confluence, JIRA, Salesforce, and more
Credal automatically keeps your connected vectors up to date in near real time
We help you manage tricky governance workflows like servicing GDPR data requests and tracking data lineage.
In practice, a lot of RAG applications need to access only a curated subset of data. Credal makes this incredible simple to achieve, both via our API and our UI
Credal’s data integration layer lets you either specify exactly how you want things like Chunking to work, or to let us manage that for your.
Credal provides an API proxy for major closed source LLMs, compatible with the OpenAI and Anthropic APIs
Named-Entity-Recognition with frictionless masking and unmasking, totally invisible to end users. Configurable per use cases
Automated audit logging, tied back to the original data source
Credal provides primitives that can be built in to user workflows for collecting feedback and corrected answers, making it easier to create fine tuned foundation models in the future. The data (and hence value) of your AI usage should accrue to you - not your vendor.
Upload your AI Acceptable Use Policies and have Credal automatically enforce them
Credal gives you everything you need to supercharge your business using generative AI, securely.
Ready to dive in?
Get Started