BigHugger
GH Repository · superduper-io

superduper

Superduper: End-to-end framework for building custom AI applications and agents.

stars
5,321
30-day movement
+10/day
Related entries
60
Connections
1
makePythonragsemantic-searchchatbotllmopspretrained-modelsaidatabasedistributed-mltorchvector-searchllm-servingtransformerspytorchmongodbpythonmldatamlopsinferencellm-inference

Superduper is an end-to-end Python framework for building custom AI applications and agents. It integrates machine learning models, including PyTorch and transformer-based ones, with databases such as MongoDB to support vector search, semantic search, and RAG workflows.

You reach for it when you want to connect pretrained models and LLM inference directly to your database for building RAG and agentic applications without gluing separate ML and data tools together.

Use it to

  • Build RAG pipelines over your database
  • Add vector and semantic search to stored data
  • Serve LLM and pretrained model inference
  • Assemble custom AI agents and chatbots
  • Run distributed ML inference in production

For Python developers building AI apps and agents

Role
rag
Language
Python
Licence
Apache-2.0
Forks
545
Open issues
29
Last push
2025-09-01
Latest release
0.0.4 · 2023-08-03
topicsragvector-searchllm-inferencemongodbpytorchmlops