Production-Grade Token Counter

Analyze:

PDF
TXT
MD
DOCX
PPTX

Features:

Real GPT token counting using tiktoken
Streaming architecture
Scanned PDF detection
Large file support
Batch processing

Upload Files (.docx, .md, .pdf, .pptx, .txt)

JSON Output

Notes

Uses real GPT tokenizer (cl100k_base)
Scanned/image-only PDFs are skipped
Optimized for very large documents
Batch processing supported