docs: OCR v2 polish on multimodal-extraction (Helm/local HF)

[kheiss-uwzoo]

Summary

Docs-only polish for multimodal-extraction.md: scannable Helm / NVCF hosted / local Hugging Face OCR v2 guidance and a clearer chart/captioning note.

Merge after #2295. Hosted-default docs assume CPU actors default to nemotron-ocr-v2 on build.nvidia.com once that eng PR lands.

Does not change releasenotes.md. The 26.05 section correctly records what 26.5.0 shipped (HF v2, Helm v1); #2239 merged after that release.

Changes

• multimodal-extraction.md: Split OCR deployment guidance (Helm / NVCF hosted / local HF); chart pdfium-path callout preserved from main; Omni caption routing note

PR scope check

• Base: main
• Files changed: 1 — docs/docs/extraction/multimodal-extraction.md
• Red flags: none

Test plan

• Spot-check OCR and chart sections in multimodal extraction
• Confirm 26.05 release notes unchanged on main
• Merge feat: switch hosted OCR CPU actor defaults to Nemotron OCR v2 #2295 first, then rebase if needed before merge

[greptile-apps]

Greptile Summary

This PR polishes the multimodal-extraction.md docs page by splitting the single OCR paragraph into three deployment-specific sections (Helm/NIM, NVCF hosted, local Hugging Face) and by replacing a vague cross-reference about chart routing with a direct statement that chart-labeled regions stay on the layout-and-OCR path.

• OCR section restructured: One paragraph is split into three clearly labeled deployment modes with matching links into the Helm README, CLI reference, and support matrix.
• Chart/captioning note tightened: The old link-only note is replaced with an explicit "not routed through the Omni caption stage" statement, reducing ambiguity for new readers.
• Forward-looking NVCF claim acknowledged: The new "Hosted remote inference (NVCF)" bullet documents behavior that depends on a companion PR (feat: switch hosted OCR CPU actor defaults to Nemotron OCR v2 #2295); the PR description explicitly calls out the required merge order.

Confidence Score: 5/5

Docs-only change with no code, config, or schema modifications; safe to merge after the documented dependency (#2295) lands.

The change touches only one documentation file. All internal cross-reference anchors resolve correctly against the current prerequisites-support-matrix.md. The one forward-looking statement (NVCF actor defaulting to the hosted OCR v2 endpoint) is gated on a separate PR that the author explicitly calls out as a merge prerequisite, so no user is mis-served while that PR is pending.

No files require special attention; the single changed file is documentation with well-validated links.

Important Files Changed

Filename	Overview
docs/docs/extraction/multimodal-extraction.md	Docs-only restructuring of OCR and chart-captioning sections; cross-reference links validate against existing anchors in prerequisites-support-matrix.md; NVCF default behavior is forward-looking (depends on #2295) but explicitly noted in the PR description.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[OCR / Scanned document] --> B{Deployment mode?}
    B -->|Helm / NIM| C[Nemotron OCR v2 NIM\nnvcr.io/nim/nvidia/nemotron-ocr-v2:1.4.0]
    B -->|Hosted NVCF\nno ocr_invoke_url set| D[build.nvidia.com\nNemotron OCR v2 endpoint\n requires #2295]
    B -->|Local Hugging Face\ngpu inference| E[Nemotron OCR v2\nmultilingual mode default]

    A2[Chart-labeled PDF region] --> F{Routing path}
    F --> G[Layout-and-OCR path\nNOT Omni caption stage]
    F -.->|optional infographic| H[Omni caption stage\ncaption_infographics=True]

Loading

%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
flowchart TD
    A[OCR / Scanned document] --> B{Deployment mode?}
    B -->|Helm / NIM| C[Nemotron OCR v2 NIM\nnvcr.io/nim/nvidia/nemotron-ocr-v2:1.4.0]
    B -->|Hosted NVCF\nno ocr_invoke_url set| D[build.nvidia.com\nNemotron OCR v2 endpoint\n requires #2295]
    B -->|Local Hugging Face\ngpu inference| E[Nemotron OCR v2\nmultilingual mode default]

    A2[Chart-labeled PDF region] --> F{Routing path}
    F --> G[Layout-and-OCR path\nNOT Omni caption stage]
    F -.->|optional infographic| H[Omni caption stage\ncaption_infographics=True]

Loading

_{Reviews (13): Last reviewed commit: "Merge branch 'main' into docs/ocr-v2-def..." | Re-trigger Greptile}