Update build-extension to allow arm compilation (#2)

Update with latest speed numbers

Chunk relational model for consistent CUDA performance

fixed batched relational

updated pipeline

updated pipeline relational to match v1

Added config to the downloaded artifacts

Pad recognizer classifier to 64-aligned output for tensor-core efficiency

Vectorize post-processing format step and default to white padding

improve install docs and process

Remove unused code: training-only methods, dead model blocks, and test files

Accept torch.Tensor inputs, remove non-release scripts

Clean up docs, make syncs optional, update examples for new modes

Make per-batch CUDA syncs and logging optional via verbose_post

Add detector_only and skip_relational inference modes

Lower default detector batch size to 8 for 16GB GPU compatibility

Reduce recognizer VRAM by discarding logits per chunk

Add pre-NMS centerness + peak filter for faster inference

updated readme.md

fix missing runtime/build deps and add Slurm helper scriptspyproject.toml: add numpy, huggingface_hub to dependencies andeditables to build-system.requires so editable installs workcleanly without manual workarounds.New scripts:- submit_install_gpu.sh: GPU-node venv creation + package build- submit_inference_viz_gpu.sh: GPU-node batched v2 inference + viz- run_ocr_inference.py: directory-level inference with JSON output

Upload config.json

removed blank config.json

updated model card

merged english and multilingual and updated model card

Duplicate from nvidia/nemotron-ocr-v2-multilingual