BigHugger
GH Repository · truefoundry

cognita

RAG (Retrieval Augmented Generation) Framework for building modular, open source applications for production by TrueFoundry

stars
4,418
30-day movement
-1-0/day
Related entries
60
Connections
1
dockerPythonllm-opsfine-tuningaideep-learningdataapplicationmachine-learningframeworkmodel-deploymentllmpythonllmopsmlopsgenerative-airagretrieval-augmented-generationtypescriptagent

Cognita is an open source RAG (Retrieval Augmented Generation) framework from TrueFoundry for building modular applications meant for production. It ships with a Docker-based toolchain and covers both Python and TypeScript components.

Use it when you want a modular, production-oriented framework for assembling RAG applications rather than building retrieval and generation plumbing from scratch.

Use it to

  • Build production RAG applications
  • Compose retrieval and generation pipelines modularly
  • Deploy RAG services with the Docker toolchain
  • Experiment with LLM retrieval-augmented workflows
  • Fine-tune or extend RAG components for your data

For Developers and ML teams building RAG applications

Role
rag
Language
Python
Licence
Apache-2.0
Forks
392
Open issues
14
Last push
2026-03-13
topicsragllmframeworkretrieval-augmented-generationmlopsproduction