deeppavlov/DeepPavlov

Health v2 · maintenance only

Adoption not part of the score

6989 stars · 1171 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

DeepPavlov is an open-source Python NLP library built on PyTorch and Hugging Face transformers for developing, training, and deploying state-of-the-art NLP models via configuration-driven pipelines. It provides pre-trained models for dialog systems, chatbots, question answering, NER, intent classification, and more, with easy REST API serving.

Use cases

• build a chatbot with python
• named entity recognition on text
• question answering over documents
• intent classification for a support bot
• slot filling for dialogue systems
• serve an NLP model as a REST API
• fine-tune BERT for NLP tasks

When to choose

• you need pre-trained NLP models (NER, QA, intent classification) without deep ML expertise
• you want configuration-driven training and inference pipelines
• you need to quickly expose an NLP model as a REST or socket API
• you are building end-to-end dialog systems or conversational assistants

When to avoid

• you need a lightweight general-purpose LLM agent framework rather than classic NLP task models
• you work outside Python or need GPU-free tiny deployments
• you want cutting-edge LLM fine-tuning features found in newer frameworks
• you need real-time streaming chat interfaces out of the box

Facets

library · maturity active

nlp machine-learning deep-learning chatbot rag llm-inference api-framework chatbots machine-learning deep-learning artificial-intelligence python windows cli conversational-ai named-entity-recognition question-answering intent-classification slot-filling bert transformers pytorch pretrained-models rest-api-serving natural-language-processing linux macos docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="deeppavlov/DeepPavlov")