# Tech Stack

**Fully extensible** 

Built on industry standard open source tech. Our RAG ingestion, indexing and execution infrastructure can be plugged into a variety of inference and embeddings models and endpoints. 

**Tech stack:**

* [LangChain](https://www.langchain.com/langchain) 

The largest community building the future of LLM apps

LangChain’s flexible abstractions and AI-first toolkit make it the #1 choice for developers when building with GenAI.

* [NextJs](https://nextjs.org/) 

The React Framework for the Web

Used by some of the world's largest companies, Next.js enables you to create **high-quality web applications** with the power of React components.

* [Vercel AI SDK](https://sdk.vercel.ai/docs)

An open source library for building AI-powered user interfaces.

The Vercel AI SDK is an open-source library designed to help developers build conversational streaming user interfaces in JavaScript and TypeScript. 

* [Supabase](https://supabase.com/docs) / Postgres 

Supabase is an open source Firebase alternative. Start your project with a Postgres database, Authentication, instant APIs, Edge Functions, Realtime subscriptions, Storage, and Vector embeddings.

* [Mistral / Mixtral](https://mistral.ai/technology/#models) models 

**Mixtral 8x7B**

Currently the best open model. A 7B sparse Mixture-of-Experts (SMoE). Uses 12B active parameters out of 45B total.

**Mistral 7B**

Our very first. A 7B transformer model, fast-deployed and easily customisable. Small, yet very powerful for a variety of use cases.

**Microsoft/WizardLM-2-8x22B**

* [PgVector](https://github.com/pgvector/pgvector) 

Open-source vector similarity search for Postgres. Store your vectors with the rest of your data. 

* [TransformersJS](https://huggingface.co/docs/transformers.js/index#transformersjs) 

Transformers.js is designed to be functionally equivalent to Hugging Face’s [transformers](https://github.com/huggingface/transformers) python library, meaning you can run the same pretrained models using a very similar API. 

* [nomic-embed-text ](https://huggingface.co/nomic-ai/nomic-embed-text-v1)

The first Open source, Open data, Open training code, Fully reproducible and auditable text embedding model with a 8192 context-length that outperforms OpenAI Ada-002 and text-embedding-3-small on both short and long context tasks.

* [Unstructured](https://unstructured.io/).io 

Unstructured effortlessly extracts and transforms complex data for use with every major vector database and LLM framework.

* [Playwright](https://playwright.dev/docs/intro) 

Playwright Test was created specifically to accommodate the needs of end-to-end testing. Playwright supports all modern rendering

* Auth with DID 

Decentralized identifiers (DIDs) are a type of globally unique identifier that enables an entity to be identified in a manner that is verifiable, persistent (as long as the DID controller desires), and does not require the use of a centralized registry.

* [Kafka](https://kafka.apache.org/)

Apache Kafka is an open-source distributed event streaming platform used by thousands of companies for high-performance data pipelines, streaming analytics, data integration, and mission-critical applications.

* [Kubernetes](https://kubernetes.io/es/) 

Kubernetes (K8s) is an open source platform to automate the implementation, scaling, and management of container applications.

* [Helm](https://helm.sh/docs/)

Helm is the package manager for Kubernetes\ <br>

**Can support:**

* custom models including openAI embeddings and inference endpoints&#x20;
* any kind of auth&#x20;
* python custom endpoints&#x20;
* custom pipelines, agents and tools&#x20;
* other vector stores

**Technicals**:

* HSNW vector indexes using PgVector 0.5&#x20;
* 50min indexing time for 1 million rows&#x20;
* Apis

<br>