| --- |
| license: apache-2.0 |
| language: |
| - en |
| tags: |
| - rag |
| - history |
| - 20th-century |
| - question-answering |
| datasets: |
| - TinyModels/chronos-wiki-corpus |
| --- |
| |
| # Chronos 🕰️ |
|
|
| A retrieval-augmented generation (RAG) bot that answers questions about 20th-century history. |
|
|
| ## What It Knows |
|
|
| - World War I & II |
| - The Cold War |
| - Major inventions and technological breakthroughs |
| - Key political events and figures |
|
|
| ## How It Works |
|
|
| Chronos searches a curated knowledge base of historical Wikipedia articles, then uses a language model to generate answers grounded in those sources. It prefers admitting ignorance over hallucinating. |
|
|
| ## Usage |
|
|
| ```python |
| from pipeline import Chronos |
| bot = Chronos("path/to/model") |
| print(bot.ask("What caused World War I?")) |
| ``` |
|
|
| The model requires a compatible LLM and tokenizer placed in the same directory. |
|
|
| ## Built With |
|
|
| - FAISS for retrieval |
| - Sentence Transformers for embeddings |
| - Cross-encoder for answer re-ranking |
|
|
| --- |
|
|
| *Part of the TinyModels collection.* |
