Sign Up | Documentation | Email List

Mixpeek listens in on changes to your database then processes each change (file_url or inline) through an inference pipeline of: extraction, generation and embedding leaving your database with fresh multimodal data always.

It removes the need of setting up architecture to track database changes, extracting content, processing and embedding it then treating each change as its' own atomic unit

We support every modality: documents, images, video, audio and of course text.

• MongoDB: https://docs.mixpeek.com/integrations/mongodb

Mixpeek is structured into two main services, each designed to handle a specific part of the process:

• API Orchestrator: Coordinates the flow between services, ensuring smooth operation and handling failures gracefully.
• Distributed Queue:
• Inference Service: Extraction, embedding, and generation of payloads

These services are containerized and can be deployed on separate servers for optimal performance and scalability.

Clone the Mixpeek repository and navigate to the SDK directory:

git clone [email protected]:mixpeek/server.git
cd server

We use poetry for all services, but there is an optional Dockerfile in each. We'll use poetry in the setup for quick deployment.

For each service you'll do the following:

1. Create a virtual environment

poetry env use python3.10

2. Activate the virtual environment

poetry shell

3. Install the requirements

poetry install

.env file:

SERVICES_CONTAINER_URL=https://localhost:8001
PYTHON_VERSION=3.11.6
OPENAI_KEY=
ENCRYPTION_KEY=

REDIS_URL=

MONGO_URL=
MONGODB_ATLAS_PUBLIC_KEY=
MONGODB_ATLAS_PRIVATE_KEY=
MONGODB_ATLAS_GROUP_ID=

AWS_ACCESS_KEY=
AWS_SECRET_KEY=
AWS_REGION=
AWS_ARN_LAMBDA=

MIXPEEK_ADMIN_TOKEN=

Then run it:

poetry run python3 -m uvicorn main:app --reload