Free
Chat with PDF
Ask questions about a PDF and get grounded answers.
Grounded answersPDF and textMultiple questionsFree
Drop or tap to upload a document
PDF, TXT, MD, HTML · max 50 MB
Frequently asked questions
How does it answer questions?
We extract the document text and ask an LLM to answer using only that content.
Can I ask multiple questions?
Yes, keep the document uploaded and ask follow-up questions.
Built with Pixeltable
Run this as tables and an API on Pixeltable Cloud.
Fact sheet
Last updated
| Modality | Document |
|---|---|
| Inputs | PDF or text file |
| Outputs | Answers grounded in the uploaded document |
| Computed column | extract_document_text, then Groq chat_completions. No embedding index |
| What updates | A new or changed row recomputes only that row. Unchanged rows stay cached. |
| Canonical doc | Computed columns |
Runnable sample
import pixeltable as pxt
from pixeltable.functions.groq import chat_completions
TableModel = pxt.model_base()
class DocChat(TableModel, name='doc_chat'):
document: pxt.Document
question: pxt.String
extracted_text = extract_document_text(document)
llm_prompt = build_doc_prompt(extracted_text, question)
answer = chat_completions(
messages=[{'role': 'user', 'content': llm_prompt}],
model='llama-3.3-70b-versatile',
).choices[0].message.contentRelated tools
- Document Summarizer
Modality: Document
Computed column: extract_document_text, then Groq chat_completions for the summary
Docs - PDF to Text Extractor
Modality: Document
Computed column: extract_document_text on the Document column
Docs