Files
DocsGPT/deployment
Alex 5a80e8cf22 docs(sandbox): correct the parsing worker's GPU note
The sentence dates from when DOCLING_OCR_ENABLED=true meant docling plus
RapidOCR. Under the new defaults OCR_ENABLED=true resolves to the native
backend with OCR_ENGINE=tesseract, a CPU subprocess that GPU libraries do not
accelerate. Name what actually uses a GPU: docling's torch-backed layout and
table models on the worker, or a DeepSeek-OCR endpoint off it.
2026-09-04 16:02:50 +01:00
..
2026-09-04 17:45:38 +04:00
2026-04-18 13:13:57 +01:00
2026-09-04 17:45:38 +04:00