#3333rd of 39 in RAG, knowledge and search
bertopic
Python topic modeling library using transformer embeddings and c-TF-IDF
- Stars
- 7.9k
- License
- MIT
- Last commit
- Aug 2026
- Last release
- Dec 2025
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
BERTopic is a Python library that groups documents into topics. By default it embeds text with sentence-transformers, reduces dimensions with UMAP, clusters with HDBSCAN, and extracts topic words with c-TF-IDF. Each step can be swapped. It also covers guided, supervised, dynamic, hierarchical, online, multimodal and zero-shot variants, plus optional LLM-generated topic labels.
Who it is for: Data scientists clustering and labeling large text collections in Python
Strengths
- Each pipeline stage (embedding, UMAP, HDBSCAN, vectorizer, representation) can be replaced or removed
- Many variants in one API: guided, supervised, online, dynamic, hierarchical, multimodal, zero-shot
- Built-in visualizations: topic map, hierarchy, heatmap, topics over time
- Representation models include KeyBERTInspired and OpenAI-based topic labeling
Weaknesses
- Library only: no web UI or server, so you write Python
- Default HDBSCAN puts many documents in outlier topic -1 that need extra handling
- Default install pulls in transformers, UMAP and HDBSCAN; lightweight setup is separate
- RAM and GPU requirements are not stated in the README
What it needs
- GPU optional
- Needs sentence-transformers, UMAP, HDBSCAN, scikit-learn
- Models: sentence-transformers, Flair, Gensim, spaCy, Universal Sentence Encoder
Also in RAG, knowledge and search
See all 39| Rank | Project | Score |
|---|---|---|
| 1 | deeptutorSelf-hosted AI tutor with knowledge bases, guided learning, books and quizzes | 82 out of 100 |
| 2 | LightRAGGraph-based RAG framework with a REST API server and web UI | 78 out of 100 |
| 3 | MorphicAI answer engine with cited search results and generative UI components | 77 out of 100 |
| #4 | MaxKBEnterprise knowledge-base agent platform with RAG, workflows and MCP tools | 75 out of 100 |
| #5 | RAGFlowRAG engine with deep document parsing, template chunking and agentic retrieval | 74 out of 100 |