58 Pull requests merged by 8 people

• Add FAQ items and create new Data page

  #593 merged Mar 15, 2025

• Remove graphviz and its dependencies from Dockerfile to reduce image size

  #579 merged Mar 15, 2025

• Another round of VDR fixes

  #592 merged Mar 15, 2025

• Add new FAQ page with 2 initial items

  #590 merged Mar 14, 2025

• Restrict max progress engines for libmode

  #587 merged Mar 14, 2025

• raise exception instead of returning none when parakeet fails

  #586 merged Mar 14, 2025

• Add example of extracting text from non-standard document types

  #584 merged Mar 14, 2025

• In pipeline_runners set is_port_in_use to use SO_REUSEADDR

  #585 merged Mar 14, 2025

• A few more edits to the new API doc

  #583 merged Mar 14, 2025

• add python api documentation

  #580 merged Mar 14, 2025

• Fix bug in TOC

  #581 merged Mar 14, 2025

• VDR changes round 3

  #578 merged Mar 14, 2025

• update environment config documentation

  #569 merged Mar 14, 2025

• add usage guide for nemoretriever-parse

  #530 merged Mar 13, 2025

• Remove use of wand/image magick

  #573 merged Mar 13, 2025

• VDR changes round 2

  #576 merged Mar 13, 2025

• Adjust openapi docs to be /docs instead of /v1/docs URL

  #574 merged Mar 13, 2025

• Make librosa optional

  #575 merged Mar 13, 2025

• remove software-properties-common

  #577 merged Mar 13, 2025

• Update python example to llama embed model

  #568 merged Mar 13, 2025

• suppress cuda related warnings

  #567 merged Mar 13, 2025

• Adjust nv-ingest wheel name to be PyPi friendly

  #571 merged Mar 13, 2025

• Update NGC API key reference in audio user guide

  #512 merged Mar 13, 2025

• adding note for parse and updated telemetry page

  #572 merged Mar 13, 2025

• updated nv-ingest name in titles and paragraphs based on feedback

  #566 merged Mar 13, 2025

• Add a pre-launch check for libmode pipeline creation

  #563 merged Mar 13, 2025

• update json output examples

  #570 merged Mar 13, 2025

• VDR Doc updates

  #565 merged Mar 13, 2025

• Add redirect for nemo/retriever/extraction/(index)

  #564 merged Mar 13, 2025

• Remove references to store task from docs and examples

  #561 merged Mar 13, 2025

• use current timestamp in telemetry when job fails

  #562 merged Mar 13, 2025

• this adds 1% to ranker recall

  #555 merged Mar 13, 2025

• add return failure mode to client

  #558 merged Mar 13, 2025

• adjust image size based on yolox-graphic-elements version

  #560 merged Mar 12, 2025

• Remove vdb_upload from supported CLI tasks.

  #559 merged Mar 12, 2025

• Updates to resolve progress engine over-subscription on low CPU-count ( less than 8 ) systems

  #557 merged Mar 12, 2025

• removed kubernetes in dev and deployment pages

  #556 merged Mar 12, 2025

• Fix sparse meta bug

  #542 merged Mar 12, 2025

• Fix broken link from quickstart to config

  #554 merged Mar 12, 2025

• 25.03 Updates including release notes and library mode

  #553 merged Mar 12, 2025

• Use full data for text encoding detection

  #552 merged Mar 12, 2025

• remove document_type suffix from source_name in image data extraction

  #549 merged Mar 11, 2025

• Add support for bmp files and treat html files as text

  #546 merged Mar 11, 2025

• Replaces ffmpeg with librosa

  #550 merged Mar 11, 2025

• fix reranker output

  #551 merged Mar 11, 2025

• Prep quickstart guide for release

  #547 merged Mar 11, 2025

• Rename index-->overview for retriever

  #545 merged Mar 11, 2025

• Fix API docs missing from TOC, add one link, fix page title wrapper

  #544 merged Mar 11, 2025

• Update Retriever overview and refactor prerequisites / support matrix

  #543 merged Mar 10, 2025

• Enable manual trigger for pypi publishing

  #541 merged Mar 10, 2025

• Conda client build fix

  #537 merged Mar 10, 2025

• fix sparse index recording for meta collection log

  #536 merged Mar 10, 2025

• add library mode support to audio extraction

  #535 merged Mar 10, 2025

• Rename field names for library mode

  #534 merged Mar 10, 2025

• add build endpoint for nemoretriever-parse

  #523 merged Mar 10, 2025

• Adjust default vlm endpoints

  #533 merged Mar 10, 2025

• Add check for environment override for graphic elements and table str…

  #532 merged Mar 9, 2025

• Reorg docs for S.S. request. First iteration

  #525 merged Mar 9, 2025

7 Pull requests opened by 5 people

• [DO NOT MERGE] Updated library mode README

  #538 opened Mar 10, 2025

• Small library mode example changes and add deployment-modes doc

  #539 opened Mar 10, 2025

• [DO NOT MERGE] updating NGC key env var for table-structure

  #540 opened Mar 10, 2025

• update python client usage notebook

  #582 opened Mar 14, 2025

• add external parkeet endpoint for library mode

  #588 opened Mar 14, 2025

• Helm chart updates for 25.3.0 Release

  #589 opened Mar 14, 2025

• Stability updates for REST service to improve very large file handling.

  #591 opened Mar 14, 2025

1 Unresolved conversation

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Refactor: Decouple Orchestration and API Layers in NV-Ingest

  #485 commented on Mar 14, 2025 • 0 new comments