docs: restructure docs directory and move fern config to fern/ (#6700)

Signed-off-by: Neal Vaidya <nealv@nvidia.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

.claude/skills/add-dynamo-docs/SKILL.md

@@ -33,11 +33,11 @@ Ask for:
 - **Page title** — appears in the sidebar and as the H1
 - **Target section** — which sidebar section (e.g., `Getting Started`, `User Guides`, `Components`)
 - **Filename** — kebab-case `.md` file (e.g., `my-new-feature.md`)
-- **Subdirectory** — which `docs/pages/` subdirectory (e.g., `getting-started`, `features`, `components`)
+- **Subdirectory** — which `docs/` subdirectory (e.g., `getting-started`, `features`, `components`)
 
 ### 2. Create the Page
 
-Create `docs/pages/<subdirectory>/<filename>.md` with SPDX header and Fern frontmatter:
+Create `docs/<subdirectory>/<filename>.md` with SPDX header and Fern frontmatter:
 
 ```markdown
 ---
@@ -53,14 +53,14 @@ title: <Page Title>
 
 ### 3. Add Navigation Entry
 
-Edit `docs/versions/dev.yml` and add the page under the correct section:
+Edit `docs/index.yml` and add the page under the correct section:
 
 ```yaml
 - page: <Page Title>
-  path: ../pages/<subdirectory>/<filename>.md
+  path: <subdirectory>/<filename>.md
 ```
 
-**Section locations in `dev.yml`** (search for the comment banner):
+**Section locations in `index.yml`** (search for the comment banner):
 - `# ==================== Getting Started ====================`
 - `# ==================== Kubernetes Deployment ====================`
 - `# ==================== User Guides ====================`
@@ -105,8 +105,6 @@ Reference images from `docs/assets/`:
 ### 5. Validate
 
 ```bash
-cd docs
-ln -sf . fern  # symlink required by Fern CLI
 fern check
 fern docs broken-links
 ```
@@ -114,7 +112,6 @@ fern docs broken-links
 ### 6. Preview Locally (Optional)
 
 ```bash
-cd docs
 fern docs dev
 ```
 
@@ -123,7 +120,8 @@ Opens a local preview at `http://localhost:3000` with hot reload. No token requi
 ### 7. Commit
 
 ```bash
-git add docs/pages/<subdirectory>/<filename>.md docs/versions/dev.yml
+git add docs/<subdirectory>/<filename>.md docs/index.yml
+
 git commit -s -m "docs: add <page-title> page"
 ```
 
@@ -131,8 +129,8 @@ git commit -s -m "docs: add <page-title> page"
 
 ### `fern check` fails
 
-- **Invalid YAML in `dev.yml`:** Check indentation — nav entries use 2-space indent. A `- page:` must be inside a `contents:` block.
-- **Missing file:** The `path:` in `dev.yml` must match the actual file location. Paths are relative to `docs/versions/` (e.g., `../pages/getting-started/quickstart.md`).
+- **Invalid YAML in `index.yml`:** Check indentation — nav entries use 2-space indent. A `- page:` must be inside a `contents:` block.
+- **Missing file:** The `path:` in `index.yml` must match the actual file location. Paths are relative to `docs/` (e.g., `getting-started/quickstart.md`).
 
 ### `fern docs broken-links` reports errors
 
@@ -143,11 +141,11 @@ git commit -s -m "docs: add <page-title> page"
 
 - **MDX parse error:** Angle-bracket URLs like `<https://example.com>` break MDX parsing. Use `[text](https://example.com)` instead.
 - **Broken links check:** The `detect_broken_links.py` job checks relative links across all docs. If your new page links to a file that doesn't exist yet, CI will fail.
-- **Fern publish error:** Check the Actions tab for the `Fern Docs` workflow. Common causes: expired `FERN_TOKEN`, invalid `docs.yml` syntax, or a file referenced in `dev.yml` that wasn't synced to `docs-website`.
+- **Fern publish error:** Check the Actions tab for the `Fern Docs` workflow. Common causes: expired `FERN_TOKEN`, invalid `fern/docs.yml` syntax, or a file referenced in `index.yml` that wasn't synced to `docs-website`.
 
 ### Page doesn't appear on the live site
 
-- **Missing nav entry:** The page exists but isn't in `docs/versions/dev.yml`. Add it.
+- **Missing nav entry:** The page exists but isn't in `docs/index.yml`. Add it.
 - **Hidden section:** The page is inside a `hidden: true` section. It's accessible by direct URL but won't appear in the sidebar.
 - **Sync delay:** After merge to `main`, the sync-dev workflow takes a few minutes to publish.
 
@@ -155,8 +153,9 @@ git commit -s -m "docs: add <page-title> page"
 
 | File | Purpose |
 |------|---------|
-| `docs/versions/dev.yml` | Navigation tree — add entries here |
-| `docs/pages/` | Content directory — create pages here |
+| `docs/index.yml` | Navigation tree — add entries here |
+| `docs/` | Content directory — create pages here |
 | `docs/assets/` | Images, SVGs, fonts |
-| `docs/convert_callouts.py` | Callout conversion rules (GitHub → Fern) |
+| `fern/docs.yml` | Fern site configuration |
+| `fern/convert_callouts.py` | Callout conversion rules (GitHub → Fern) |
 | `docs/README.md` | Full architecture guide |

.claude/skills/rm-dynamo-docs/SKILL.md

@@ -30,25 +30,25 @@ Must be in the `dynamo` repo (not `dynamo-tpm`). Architecture details: `docs/REA
 ### 1. Identify the Page
 
 Ask for the page to remove (accepts any of):
-- File path (e.g., `docs/pages/guides/old-page.md`)
+- File path (e.g., `docs/guides/old-page.md`)
 - Page title (e.g., "Old Page")
 - Topic keyword to search for
 
 If given a title or keyword, search for the page:
 ```bash
 # Search by title in navigation
-grep -n "<title>" docs/versions/dev.yml
+grep -n "<title>" docs/index.yml
 
 # Search by keyword in content
-grep -rl "<keyword>" docs/pages/
+grep -rl "<keyword>" docs/
 ```
 
 ### 2. Find the Navigation Entry
 
-Locate the page's entry in `docs/versions/dev.yml`:
+Locate the page's entry in `docs/index.yml`:
 
 ```bash
-grep -n "<filename>" docs/versions/dev.yml
+grep -n "<filename>" docs/index.yml
 ```
 
 Note the exact `- page:` block and its indentation level. If the page is the
@@ -60,10 +60,10 @@ Search for references to this page from other docs:
 
 ```bash
 # Search for the filename across all docs pages
-grep -r "<filename>" docs/pages/ --include="*.md"
+grep -r "<filename>" docs/ --include="*.md"
 
 # Also check the navigation file for any cross-references
-grep -r "<filename>" docs/versions/
+grep -r "<filename>" docs/
 ```
 
 Report any files that link to the page being removed — these links will break
@@ -72,12 +72,12 @@ and need updating.
 ### 4. Remove the Markdown File
 
 ```bash
-git rm docs/pages/<subdirectory>/<filename>.md
+git rm docs/<subdirectory>/<filename>.md
 ```
 
 ### 5. Remove the Navigation Entry
 
-Edit `docs/versions/dev.yml` and delete the `- page:` block (and its `path:`
+Edit `docs/index.yml` and delete the `- page:` block (and its `path:`
 line). If this was the last page in a section, remove the entire `- section:`
 block.
 
@@ -91,8 +91,6 @@ For each file that linked to the removed page:
 ### 7. Validate
 
 ```bash
-cd docs
-ln -sf . fern  # symlink required by Fern CLI
 fern check
 fern docs broken-links
 ```
@@ -100,7 +98,6 @@ fern docs broken-links
 ### 8. Preview Locally (Optional)
 
 ```bash
-cd docs
 fern docs dev
 ```
 
@@ -117,18 +114,18 @@ git commit -s -m "docs: remove <page-title> page"
 
 ### `fern check` fails
 
-- **Orphaned nav entry:** You deleted the file but left the `- page:` entry in `dev.yml`. Remove it.
-- **Empty section:** If the removed page was the only entry in a section, delete the entire `- section:` block from `dev.yml`.
+- **Orphaned nav entry:** You deleted the file but left the `- page:` entry in `index.yml`. Remove it.
+- **Empty section:** If the removed page was the only entry in a section, delete the entire `- section:` block from `index.yml`.
 
 ### `fern docs broken-links` reports errors
 
-- **Incoming links to removed page:** Other pages still link to the deleted file. Search with `grep -r "<filename>" docs/pages/` and update or remove those links.
+- **Incoming links to removed page:** Other pages still link to the deleted file. Search with `grep -r "<filename>" docs/` and update or remove those links.
 
 ### CI fails after merge
 
 - **MDX parse error:** Angle-bracket URLs like `<https://example.com>` break MDX parsing. Use `[text](https://example.com)` instead.
 - **Broken links check:** The `detect_broken_links.py` job found pages that still reference the removed file. Fix all incoming links before merging.
-- **Fern publish error:** Check the Actions tab for the `Fern Docs` workflow. Common causes: expired `FERN_TOKEN`, invalid `docs.yml` syntax.
+- **Fern publish error:** Check the Actions tab for the `Fern Docs` workflow. Common causes: expired `FERN_TOKEN`, invalid `fern/docs.yml` syntax.
 
 ### Page still appears on the live site
 
@@ -139,6 +136,7 @@ git commit -s -m "docs: remove <page-title> page"
 
 | File | Purpose |
 |------|---------|
-| `docs/versions/dev.yml` | Navigation tree — remove entries here |
-| `docs/pages/` | Content directory — delete pages here |
+| `docs/index.yml` | Navigation tree — remove entries here |
+| `docs/` | Content directory — delete pages here |
+| `fern/docs.yml` | Fern site configuration |
 | `docs/README.md` | Full architecture guide |

.claude/skills/update-dynamo-docs/SKILL.md

@@ -30,29 +30,29 @@ Must be in the `dynamo` repo (not `dynamo-tpm`). Architecture details: `docs/REA
 ### 1. Locate the Page
 
 Ask for the page to update (accepts any of):
-- File path (e.g., `docs/pages/guides/quickstart.md`)
+- File path (e.g., `docs/guides/quickstart.md`)
 - Page title (e.g., "Quickstart")
 - Topic keyword to search for
 
 If given a title or keyword, find the page:
 ```bash
 # Search by title in navigation
-grep -n "<title>" docs/versions/dev.yml
+grep -n "<title>" docs/index.yml
 
 # Search by keyword in content
-grep -rl "<keyword>" docs/pages/
+grep -rl "<keyword>" docs/
 ```
 
 ### 2. Read Current Content
 
 Read the page and its navigation entry:
-- The markdown file in `docs/pages/`
-- The corresponding entry in `docs/versions/dev.yml`
+- The markdown file in `docs/`
+- The corresponding entry in `docs/index.yml`
 
 Note the current:
 - **Title** (from frontmatter `title:` field)
 - **Section** (which sidebar section it belongs to)
-- **Path** (relative path in `dev.yml`)
+- **Path** (relative path in `index.yml`)
 
 ### 3. Apply Edits
 
@@ -65,16 +65,16 @@ Edit the markdown file directly. No navigation changes needed.
 #### Title Change
 
 1. Update the `title:` field in the markdown frontmatter
-2. Update the `- page:` display name in `docs/versions/dev.yml`
+2. Update the `- page:` display name in `docs/index.yml`
 
 #### Section Move
 
 1. Move the markdown file to the new subdirectory:
    ```bash
-   git mv docs/pages/<old-subdir>/<file>.md docs/pages/<new-subdir>/<file>.md
+   git mv docs/<old-subdir>/<file>.md docs/<new-subdir>/<file>.md
    ```
-2. Remove the old `- page:` entry from `dev.yml`
-3. Add a new `- page:` entry under the target section in `dev.yml`
+2. Remove the old `- page:` entry from `index.yml`
+3. Add a new `- page:` entry under the target section in `index.yml`
 4. Update the `path:` to reflect the new location
 
 ### 4. Check Incoming Links (If Path Changed)
@@ -83,8 +83,8 @@ If the file was moved, search for references that need updating:
 
 ```bash
 # Search for the old path across all docs
-grep -r "<old-filename>" docs/pages/ --include="*.md"
-grep -r "<old-filename>" docs/versions/
+grep -r "<old-filename>" docs/ --include="*.md"
+grep -r "<old-filename>" docs/
 ```
 
 Update all references to point to the new path.
@@ -114,8 +114,6 @@ Use standard **GitHub-flavored markdown**. For callouts, use GitHub's native syn
 ### 6. Validate
 
 ```bash
-cd docs
-ln -sf . fern  # symlink required by Fern CLI
 fern check
 fern docs broken-links
 ```
@@ -123,7 +121,6 @@ fern docs broken-links
 ### 7. Preview Locally (Optional)
 
 ```bash
-cd docs
 fern docs dev
 ```
 
@@ -132,7 +129,7 @@ Opens a local preview at `http://localhost:3000` with hot reload. No token requi
 ### 8. Commit
 
 ```bash
-git add docs/pages/ docs/versions/dev.yml
+git add docs/
 git commit -s -m "docs: update <page-title>"
 ```
 
@@ -140,32 +137,33 @@ git commit -s -m "docs: update <page-title>"
 
 ### `fern check` fails
 
-- **Invalid YAML in `dev.yml`:** Check indentation — nav entries use 2-space indent. A `- page:` must be inside a `contents:` block.
-- **Missing file:** If you moved a page, the `path:` in `dev.yml` must match the new location.
-- **Duplicate entry:** The same page appears twice in `dev.yml`. Remove the old entry after a section move.
+- **Invalid YAML in `index.yml`:** Check indentation — nav entries use 2-space indent. A `- page:` must be inside a `contents:` block.
+- **Missing file:** If you moved a page, the `path:` in `index.yml` must match the new location.
+- **Duplicate entry:** The same page appears twice in `index.yml`. Remove the old entry after a section move.
 
 ### `fern docs broken-links` reports errors
 
-- **Stale internal links:** After moving or renaming a page, other pages may still link to the old path. Search with `grep -r "<old-filename>" docs/pages/` and update references.
+- **Stale internal links:** After moving or renaming a page, other pages may still link to the old path. Search with `grep -r "<old-filename>" docs/` and update references.
 - **Anchor not found:** A `#section-heading` link doesn't match any heading in the target page. Check if the heading text changed.
 
 ### CI fails after merge
 
 - **MDX parse error:** Angle-bracket URLs like `<https://example.com>` break MDX parsing. Use `[text](https://example.com)` instead.
 - **Broken links check:** The `detect_broken_links.py` job found stale references to the old path. Fix all incoming links before merging.
-- **Fern publish error:** Check the Actions tab for the `Fern Docs` workflow. Common causes: expired `FERN_TOKEN`, invalid `docs.yml` syntax, or a moved file that wasn't synced to `docs-website`.
+- **Fern publish error:** Check the Actions tab for the `Fern Docs` workflow. Common causes: expired `FERN_TOKEN`, invalid `fern/docs.yml` syntax, or a moved file that wasn't synced to `docs-website`.
 
 ### Changes don't appear on the live site
 
-- **Title mismatch:** You updated the frontmatter `title:` but not the `- page:` name in `dev.yml` (or vice versa). Keep them in sync.
+- **Title mismatch:** You updated the frontmatter `title:` but not the `- page:` name in `index.yml` (or vice versa). Keep them in sync.
 - **Sync delay:** After merge to `main`, the sync-dev workflow takes a few minutes to publish.
 
 ## Key References
 
 | File | Purpose |
 |------|---------|
-| `docs/versions/dev.yml` | Navigation tree — update entries here if title/path changes |
-| `docs/pages/` | Content directory — edit pages here |
+| `docs/index.yml` | Navigation tree — update entries here if title/path changes |
+| `docs/` | Content directory — edit pages here |
 | `docs/assets/` | Images, SVGs, fonts |
-| `docs/convert_callouts.py` | Callout conversion rules (GitHub → Fern) |
+| `fern/docs.yml` | Fern site configuration |
+| `fern/convert_callouts.py` | Callout conversion rules (GitHub → Fern) |
 | `docs/README.md` | Full architecture guide |

.github/filters.yaml

@@ -100,7 +100,7 @@ operator:
   - *ci
   - 'deploy/operator/**'
   - 'deploy/operator/.*'
-  - 'docs/pages/kubernetes/api-reference.md'
+  - 'docs/kubernetes/api-reference.md'
 
 deploy:
   - '!**/*.md'

.github/workflows/fern-docs.yml

@@ -101,13 +101,7 @@ jobs:
       - name: Install Fern CLI
         run: npm install -g fern-api
 
-      # Fern CLI expects a fern/ directory; create symlink after docs/ migration
-      - name: Create fern directory symlink
-        working-directory: docs
-        run: ln -s . fern
-
       - name: Validate Fern configuration
-        working-directory: docs
         run: fern check
 
   fern-broken-links:
@@ -130,13 +124,7 @@ jobs:
       - name: Install Fern CLI
         run: npm install -g fern-api
 
-      # Fern CLI expects a fern/ directory; create symlink after docs/ migration
-      - name: Create fern directory symlink
-        working-directory: docs
-        run: ln -s . fern
-
       - name: Check for broken links
-        working-directory: docs
         run: fern docs broken-links
 
   #############################################################################
@@ -178,23 +166,34 @@ jobs:
           token: ${{ secrets.GITHUB_TOKEN }}
       - name: Sync dev content from main
         run: |
-          # Sync pages/ directory (dev content)
-          echo "Syncing pages/ to docs-website branch..."
-          rm -rf docs-checkout/fern/pages
-          cp -r source-checkout/docs/pages docs-checkout/fern/pages
-
-          # Sync versions/dev.yml (dev navigation)
-          echo "Syncing versions/dev.yml to docs-website branch..."
-          cp source-checkout/docs/versions/dev.yml docs-checkout/fern/versions/dev.yml
+          # Sync content directories from docs/ to fern/pages-dev/ on docs-website
+          # Excludes non-content items that are synced separately or not needed
+          echo "Syncing content pages to docs-website branch..."
+          rm -rf docs-checkout/fern/pages-dev
+          mkdir -p docs-checkout/fern/pages-dev
+          rsync -a \
+            --exclude='assets' \
+            --exclude='blogs' \
+            --exclude='diagrams' \
+            --exclude='index.yml' \
+            source-checkout/docs/ docs-checkout/fern/pages-dev/
+
+          # Sync index.yml as versions/dev.yml and transform paths for docs-website layout
+          echo "Syncing index.yml to docs-website branch as versions/dev.yml..."
+          cp source-checkout/docs/index.yml docs-checkout/fern/versions/dev.yml
 
           # Sync assets/ directory
           echo "Syncing assets/ to docs-website branch..."
           rm -rf docs-checkout/fern/assets
           cp -r source-checkout/docs/assets docs-checkout/fern/assets
 
+          # Symlink assets into pages-dev/ so relative image paths in markdown resolve correctly
+          # (e.g. ../../assets/img/foo.png from pages-dev/observability/metrics.md)
+          ln -sfn ../assets docs-checkout/fern/pages-dev/assets
+
           # Sync fern.config.json
           echo "Syncing fern.config.json to docs-website branch..."
-          cp source-checkout/docs/fern.config.json docs-checkout/fern/fern.config.json
+          cp source-checkout/fern/fern.config.json docs-checkout/fern/fern.config.json
 
           # Sync docs/README.md (developer guide, referenced from dev.yml)
           if [ -f source-checkout/docs/README.md ]; then
@@ -202,20 +201,20 @@ jobs:
           fi
 
           # Sync .gitignore if it exists
-          if [ -f source-checkout/docs/.gitignore ]; then
-            cp source-checkout/docs/.gitignore docs-checkout/fern/.gitignore
+          if [ -f source-checkout/fern/.gitignore ]; then
+            cp source-checkout/fern/.gitignore docs-checkout/fern/.gitignore
           fi
 
           # Sync convert_callouts.py script
-          if [ -f source-checkout/docs/convert_callouts.py ]; then
-            cp source-checkout/docs/convert_callouts.py docs-checkout/fern/convert_callouts.py
+          if [ -f source-checkout/fern/convert_callouts.py ]; then
+            cp source-checkout/fern/convert_callouts.py docs-checkout/fern/convert_callouts.py
           fi
 
           # Sync components/ directory (e.g., CustomFooter.tsx)
-          if [ -d source-checkout/docs/components ]; then
+          if [ -d source-checkout/fern/components ]; then
             echo "Syncing components/ to docs-website branch..."
             rm -rf docs-checkout/fern/components
-            cp -r source-checkout/docs/components docs-checkout/fern/components
+            cp -r source-checkout/fern/components docs-checkout/fern/components
           fi
 
           # Sync blogs/ directory
@@ -226,15 +225,26 @@ jobs:
           fi
 
           # Sync main.css
-          if [ -f source-checkout/docs/main.css ]; then
+          if [ -f source-checkout/fern/main.css ]; then
             echo "Syncing main.css to docs-website branch..."
-            cp source-checkout/docs/main.css docs-checkout/fern/main.css
+            cp source-checkout/fern/main.css docs-checkout/fern/main.css
           fi
 
+      - name: Transform paths in dev.yml for docs-website layout
+        run: |
+          # In the source repo, index.yml uses paths relative to docs/ (e.g. getting-started/quickstart.md).
+          # On docs-website, fern/versions/dev.yml needs ../pages-dev/ prefix for content
+          # and ../ prefix for blogs (which are a sibling of pages/, not inside it).
+          #
+          # Order matters: transform blogs/ first so the second sub() won't re-match them
+          # (they'll start with ../ which doesn't match ^[a-zA-Z]).
+          yq -i '(.. | select(has("path")).path) |= sub("^blogs/", "../blogs/")' docs-checkout/fern/versions/dev.yml
+          yq -i '(.. | select(has("path")).path) |= sub("^([a-zA-Z])", "../pages-dev/${1}")' docs-checkout/fern/versions/dev.yml
+
       - name: Convert GitHub callouts to Fern format
         run: |
           echo "Converting GitHub-style callouts to Fern format in pages/..."
-          python3 docs-checkout/fern/convert_callouts.py --dir docs-checkout/fern/pages
+          python3 docs-checkout/fern/convert_callouts.py --dir docs-checkout/fern/pages-dev
           echo "Callout conversion complete."
 
       - name: Update docs.yml preserving products
@@ -248,7 +258,11 @@ jobs:
           cat /tmp/preserved_product.yml
 
           # Copy docs.yml from source to get config updates (redirects, layout, etc.)
-          cp ../../source-checkout/docs/docs.yml docs.yml
+          cp ../../source-checkout/fern/docs.yml docs.yml
+
+          # Fix asset paths: source uses ../docs/assets/ (relative to fern/ in source repo),
+          # but on docs-website assets live at fern/assets/ so we need ./assets/
+          sed -i 's|\.\./docs/assets/|./assets/|g' docs.yml
 
           # Restore the preserved products[0] block
           yq -i '.products[0] = load("/tmp/preserved_product.yml")' docs.yml
@@ -290,11 +304,15 @@ jobs:
         env:
           FERN_TOKEN: ${{ secrets.FERN_TOKEN }}
         run: |
-          OUTPUT=$(fern generate --docs --preview 2>&1)
+          if OUTPUT=$(fern generate --docs --preview 2>&1); then
+            FERN_EXIT=0
+          else
             FERN_EXIT=$?
+          fi
           echo "$OUTPUT"
           if [ $FERN_EXIT -ne 0 ]; then
-            echo "::warning::Fern docs preview generation failed (exit $FERN_EXIT)"
+            echo "::error::Fern docs preview generation failed (exit $FERN_EXIT)"
+            exit 1
           fi
           URL=$(echo "$OUTPUT" | grep -oP 'Published docs to \K\S+') || true
           if [ -n "$URL" ]; then
@@ -411,10 +429,10 @@ jobs:
         run: |
           TAG="${{ steps.version.outputs.tag }}"
 
-          echo "Creating fern/pages-$TAG/ from fern/pages/..."
+          echo "Creating fern/pages-$TAG/ from fern/pages-dev/..."
 
-          # Copy current pages/ to pages-vX.Y.Z/
-          cp -r fern/pages "fern/pages-$TAG"
+          # Copy current pages-dev/ to pages-vX.Y.Z/
+          cp -r fern/pages-dev "fern/pages-$TAG"
 
           echo "Created fern/pages-$TAG/"
           ls -la "fern/pages-$TAG/" | head -20
@@ -466,8 +484,8 @@ jobs:
           sed -i "s/# Navigation structure for Latest version/# Navigation structure for $TAG version/" "$VERSION_FILE"
           sed -i "s|# Matching https://docs.nvidia.com/dynamo/latest/|# Snapshot from tag $TAG|" "$VERSION_FILE"
 
-          # Update all page paths from ../pages/ to ../pages-vX.Y.Z/
-          sed -i "s|path: \.\./pages/|path: ../pages-$TAG/|g" "$VERSION_FILE"
+          # Update all page paths from ../pages-dev/ to ../pages-vX.Y.Z/
+          sed -i "s|path: \.\./pages-dev/|path: ../pages-$TAG/|g" "$VERSION_FILE"
 
           echo "Created $VERSION_FILE"
           echo "First 30 lines:"

CONTRIBUTING.md

@@ -129,7 +129,7 @@ Issues labeled `good-first-issue` are sized for new contributors. We provide ext
 
 <!-- We were given the feedback that having information on architecture, languages used, etc. would be helpful for external contributors -->
 
-Understanding Dynamo's architecture helps you find where to make changes. For the complete picture, see the [Architecture Documentation](docs/pages/design-docs/architecture.md) and [Support Matrix](docs/pages/reference/support-matrix.md).
+Understanding Dynamo's architecture helps you find where to make changes. For the complete picture, see the [Architecture Documentation](docs/design-docs/architecture.md) and [Support Matrix](docs/reference/support-matrix.md).
 
 ![Dynamo Architecture](docs/assets/img/architecture.png)
 
@@ -148,9 +148,9 @@ Understanding Dynamo's architecture helps you find where to make changes. For th
 
 | Plane | Purpose | Documentation |
 |-------|---------|---------------|
-| **Discovery Plane** | Service registration and discovery across components | [docs/pages/design-docs/distributed-runtime.md](docs/pages/design-docs/distributed-runtime.md) |
-| **Request Plane** | High-performance request routing between components | [docs/pages/design-docs/request-plane.md](docs/pages/design-docs/request-plane.md) |
-| **KV Event Plane** | KV cache event propagation for cache-aware routing | [docs/pages/design-docs/event-plane.md](docs/pages/design-docs/event-plane.md) |
+| **Discovery Plane** | Service registration and discovery across components | [docs/design-docs/distributed-runtime.md](docs/design-docs/distributed-runtime.md) |
+| **Request Plane** | High-performance request routing between components | [docs/design-docs/request-plane.md](docs/design-docs/request-plane.md) |
+| **KV Event Plane** | KV cache event propagation for cache-aware routing | [docs/design-docs/event-plane.md](docs/design-docs/event-plane.md) |
 
 ### Kubernetes Deployment
 

README.md

@@ -22,7 +22,7 @@ limitations under the License.
 [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/ai-dynamo/dynamo)
 [![Discord](https://dcbadge.limes.pink/api/server/D92uqZRjCZ?style=flat)](https://discord.gg/D92uqZRjCZ) ![Community Contributors](https://img.shields.io/badge/community_contributors-70%2B-brightgreen)
 
-| **[Roadmap](https://github.com/ai-dynamo/dynamo/issues/5506)** | **[Support Matrix](https://github.com/ai-dynamo/dynamo/blob/main/docs/pages/reference/support-matrix.md)** | **[Docs](https://docs.nvidia.com/dynamo/)** | **[Recipes](https://github.com/ai-dynamo/dynamo/tree/main/recipes)** | **[Examples](https://github.com/ai-dynamo/dynamo/tree/main/examples)** | **[Prebuilt Containers](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-dynamo/collections/ai-dynamo)** | **[Design Proposals](https://github.com/ai-dynamo/enhancements)** | **[Blogs](https://developer.nvidia.com/blog/tag/nvidia-dynamo)**
+| **[Roadmap](https://github.com/ai-dynamo/dynamo/issues/5506)** | **[Support Matrix](https://github.com/ai-dynamo/dynamo/blob/main/docs/reference/support-matrix.md)** | **[Docs](https://docs.nvidia.com/dynamo/)** | **[Recipes](https://github.com/ai-dynamo/dynamo/tree/main/recipes)** | **[Examples](https://github.com/ai-dynamo/dynamo/tree/main/examples)** | **[Prebuilt Containers](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-dynamo/collections/ai-dynamo)** | **[Design Proposals](https://github.com/ai-dynamo/enhancements)** | **[Blogs](https://developer.nvidia.com/blog/tag/nvidia-dynamo)**
 
 # NVIDIA Dynamo
 
@@ -48,17 +48,17 @@ Built in Rust for performance and Python for extensibility, Dynamo is fully open
 
 ## Backend Feature Support
 
-| | [SGLang](docs/pages/backends/sglang/README.md) | [TensorRT-LLM](docs/pages/backends/trtllm/README.md) | [vLLM](docs/pages/backends/vllm/README.md) |
+| | [SGLang](docs/backends/sglang/README.md) | [TensorRT-LLM](docs/backends/trtllm/README.md) | [vLLM](docs/backends/vllm/README.md) |
 |---|:----:|:----------:|:--:|
 | **Best For** | High-throughput serving | Maximum performance | Broadest feature coverage |
-| [**Disaggregated Serving**](docs/pages/design-docs/disagg-serving.md) | ✅ | ✅ | ✅ |
-| [**KV-Aware Routing**](docs/pages/components/router/README.md) | ✅ | ✅ | ✅ |
-| [**SLA-Based Planner**](docs/pages/components/planner/planner-guide.md) | ✅ | ✅ | ✅ |
-| [**KVBM**](docs/pages/components/kvbm/README.md) | 🚧 | ✅ | ✅ |
-| [**Multimodal**](docs/pages/features/multimodal/README.md) | ✅ | ✅ | ✅ |
-| [**Tool Calling**](docs/pages/agents/tool-calling.md) | ✅ | ✅ | ✅ |
+| [**Disaggregated Serving**](docs/design-docs/disagg-serving.md) | ✅ | ✅ | ✅ |
+| [**KV-Aware Routing**](docs/components/router/README.md) | ✅ | ✅ | ✅ |
+| [**SLA-Based Planner**](docs/components/planner/planner-guide.md) | ✅ | ✅ | ✅ |
+| [**KVBM**](docs/components/kvbm/README.md) | 🚧 | ✅ | ✅ |
+| [**Multimodal**](docs/features/multimodal/README.md) | ✅ | ✅ | ✅ |
+| [**Tool Calling**](docs/agents/tool-calling.md) | ✅ | ✅ | ✅ |
 
-> **[Full Feature Matrix →](docs/pages/reference/feature-matrix.md)** — Detailed compatibility including LoRA, Request Migration, Speculative Decoding, and feature interactions.
+> **[Full Feature Matrix →](docs/reference/feature-matrix.md)** — Detailed compatibility including LoRA, Request Migration, Speculative Decoding, and feature interactions.
 
 ## Dynamo Architecture
 
@@ -66,7 +66,7 @@ Built in Rust for performance and Python for extensibility, Dynamo is fully open
   <img src="./docs/assets/img/frontpage-architecture.png" alt="Dynamo architecture" width="600" />
 </p>
 
-> **[Architecture Deep Dive →](docs/pages/design-docs/architecture.md)**
+> **[Architecture Deep Dive →](docs/design-docs/architecture.md)**
 
 ## Latest News
 
@@ -87,7 +87,7 @@ Want to help shape the future of distributed LLM inference? See the **[Contribut
 # Local Quick Start
 
 The following examples require a few system level packages.
-Recommended to use Ubuntu 24.04 with a x86_64 CPU. See [docs/pages/reference/support-matrix.md](docs/pages/reference/support-matrix.md)
+Recommended to use Ubuntu 24.04 with a x86_64 CPU. See [docs/reference/support-matrix.md](docs/reference/support-matrix.md)
 
 ## Install Dynamo
 
@@ -108,7 +108,7 @@ docker run --gpus all --network host --rm -it nvcr.io/nvidia/ai-dynamo/vllm-runt
 
 > **Tip:** To run frontend and worker in the same container, either run processes in background with `&` (see below), or open a second terminal and use `docker exec -it <container_id> bash`.
 
-See [Release Artifacts](docs/pages/reference/release-artifacts.md#container-images) for available versions.
+See [Release Artifacts](docs/reference/release-artifacts.md#container-images) for available versions.
 
 ### Option B: Install from PyPI
 
@@ -143,7 +143,7 @@ pip install torch==2.9.0 torchvision --index-url https://download.pytorch.org/wh
 pip install --pre --extra-index-url https://pypi.nvidia.com "ai-dynamo[trtllm]"
 ```
 
-> **Note:** TensorRT-LLM requires `pip` due to a transitive Git URL dependency that `uv` doesn't resolve. We recommend using the [TensorRT-LLM container](docs/pages/reference/release-artifacts.md#container-images) for broader compatibility.
+> **Note:** TensorRT-LLM requires `pip` due to a transitive Git URL dependency that `uv` doesn't resolve. We recommend using the [TensorRT-LLM container](docs/reference/release-artifacts.md#container-images) for broader compatibility.
 
 **vLLM**
 
@@ -222,7 +222,7 @@ For production deployments on Kubernetes clusters with multiple GPUs.
 ## Prerequisites
 
 - Kubernetes cluster with GPU nodes
-- [Dynamo Platform installed](docs/pages/kubernetes/README.md)
+- [Dynamo Platform installed](docs/kubernetes/README.md)
 - HuggingFace token for model downloads
 
 ## Production Recipes
@@ -348,8 +348,8 @@ python3 -m dynamo.frontend
 
 Dynamo provides comprehensive benchmarking tools:
 
-- **[Benchmarking Guide](docs/pages/benchmarks/benchmarking.md)** – Compare deployment topologies using AIPerf
-- **[SLA-Driven Deployments](docs/pages/components/planner/planner-guide.md)** – Optimize deployments to meet SLA requirements
+- **[Benchmarking Guide](docs/benchmarks/benchmarking.md)** – Compare deployment topologies using AIPerf
+- **[SLA-Driven Deployments](docs/components/planner/planner-guide.md)** – Optimize deployments to meet SLA requirements
 
 ## Frontend OpenAPI Specification
 
@@ -359,11 +359,11 @@ The OpenAI-compatible frontend exposes an OpenAPI 3 spec at `/openapi.json`. To
 cargo run -p dynamo-llm --bin generate-frontend-openapi
 ```
 
-This writes to `docs/pages/reference/api/openapi.json`.
+This writes to `docs/reference/api/openapi.json`.
 
 ## Service Discovery and Messaging
 
-Dynamo uses TCP for inter-component communication. On Kubernetes, native resources ([CRDs + EndpointSlices](docs/pages/kubernetes/service-discovery.md)) handle service discovery. External services are optional for most deployments:
+Dynamo uses TCP for inter-component communication. On Kubernetes, native resources ([CRDs + EndpointSlices](docs/kubernetes/service-discovery.md)) handle service discovery. External services are optional for most deployments:
 
 | Deployment | etcd | NATS | Notes |
 |------------|------|------|-------|
@@ -389,11 +389,11 @@ See [TRT-LLM on Slurm](examples/basics/multinode/trtllm/README.md) for deploymen
 - [10/16] [How Baseten achieved 2x faster inference with NVIDIA Dynamo](https://www.baseten.co/blog/how-baseten-achieved-2x-faster-inference-with-nvidia-dynamo/)
 
 <!-- Reference links for Feature Compatibility Matrix -->
-[disagg]: docs/pages/design-docs/disagg-serving.md
-[kv-routing]: docs/pages/components/router/README.md
-[planner]: docs/pages/components/planner/planner-guide.md
-[kvbm]: docs/pages/components/kvbm/README.md
+[disagg]: docs/design-docs/disagg-serving.md
+[kv-routing]: docs/components/router/README.md
+[planner]: docs/components/planner/planner-guide.md
+[kvbm]: docs/components/kvbm/README.md
 [mm]: examples/multimodal/
-[migration]: docs/pages/fault-tolerance/request-migration.md
+[migration]: docs/fault-tolerance/request-migration.md
 [lora]: examples/backends/vllm/deploy/lora/README.md
-[tools]: docs/pages/agents/tool-calling.md
+[tools]: docs/agents/tool-calling.md

benchmarks/README.md

@@ -20,7 +20,7 @@ This directory contains benchmarking scripts and tools for performance evaluatio
 ## Quick Start
 
 ### Benchmark a Dynamo Deployment
-First, deploy your DynamoGraphDeployment using the [deployment documentation](../docs/pages/kubernetes/), then:
+First, deploy your DynamoGraphDeployment using the [deployment documentation](../docs/kubernetes/), then:
 
 ```bash
 # Port-forward your deployment to http://localhost:8000
@@ -71,4 +71,4 @@ Detailed information is provided in the `prefix_data_generator` directory.
 
 ## Comprehensive Guide
 
-For detailed documentation, configuration options, and advanced usage, see the [complete benchmarking guide](../docs/pages/benchmarks/benchmarking.md).
+For detailed documentation, configuration options, and advanced usage, see the [complete benchmarking guide](../docs/benchmarks/benchmarking.md).

benchmarks/router/README.md

@@ -165,7 +165,7 @@ To see all available router arguments, run:
 python -m dynamo.frontend --help
 ```
 
-For detailed explanations of router arguments (especially KV cache routing parameters), see the [Router Guide](../../docs/pages/components/router/router-guide.md).
+For detailed explanations of router arguments (especially KV cache routing parameters), see the [Router Guide](../../docs/components/router/router-guide.md).
 
 > [!Note]
 > If you're unsure whether your backend engines correctly emit KV events for certain models (e.g., hybrid models like gpt-oss or nemotron nano 2), use the `--no-kv-events` flag to disable KV event tracking and use approximate KV indexing instead:
@@ -184,7 +184,7 @@ When you launch prefill workers using `run_engines.sh --prefill`, the frontend a
 - Uses the same routing mode as the frontend's `--router-mode` setting
 - Seamlessly integrates with your decode workers for token generation
 
-No additional configuration is needed - simply launch both decode and prefill workers, and the system handles the rest. See the [Router Guide](../../docs/pages/components/router/router-guide.md#disaggregated-serving) for more details.
+No additional configuration is needed - simply launch both decode and prefill workers, and the system handles the rest. See the [Router Guide](../../docs/components/router/router-guide.md#disaggregated-serving) for more details.
 
 > [!Note]
 > The unified frontend with automatic prefill routing is currently enabled for vLLM and TensorRT-LLM backends. For SGLang (work in progress), you need to launch a separate standalone router as the prefill router targeting the prefill endpoints. See example script: [`examples/backends/sglang/launch/disagg_router.sh`](../../examples/backends/sglang/launch/disagg_router.sh)

components/README.md

@@ -25,9 +25,9 @@ This directory contains the core components that make up the Dynamo inference fr
 
 Dynamo supports multiple inference engines, each with their own deployment configurations and capabilities:
 
-- **[vLLM](/docs/pages/backends/vllm/README.md)** - Full-featured vLLM integration with disaggregated serving, KV-aware routing, SLA-based planning, native KV cache events, and NIXL-based transfer mechanisms
-- **[SGLang](/docs/pages/backends/sglang/README.md)** - SGLang engine integration with ZMQ-based communication, supporting disaggregated serving and KV-aware routing
-- **[TensorRT-LLM](/docs/pages/backends/trtllm/README.md)** - TensorRT-LLM integration with disaggregated serving capabilities and TensorRT acceleration
+- **[vLLM](/docs/backends/vllm/README.md)** - Full-featured vLLM integration with disaggregated serving, KV-aware routing, SLA-based planning, native KV cache events, and NIXL-based transfer mechanisms
+- **[SGLang](/docs/backends/sglang/README.md)** - SGLang engine integration with ZMQ-based communication, supporting disaggregated serving and KV-aware routing
+- **[TensorRT-LLM](/docs/backends/trtllm/README.md)** - TensorRT-LLM integration with disaggregated serving capabilities and TensorRT acceleration
 
 Each engine provides launch and deploy scripts for different deployment patterns in the [examples](../examples/backends/) folder.
 

components/src/dynamo/frontend/README.md

@@ -5,4 +5,4 @@
 
 The API gateway for serving LLM inference requests with OpenAI-compatible HTTP and KServe gRPC endpoints.
 
-See [docs/pages/components/frontend/](../../../../docs/pages/components/frontend/) for documentation.
+See [docs/components/frontend/](../../../../docs/components/frontend/) for documentation.

components/src/dynamo/mocker/README.md

@@ -67,7 +67,7 @@ python -m dynamo.mocker \
 
 The profile results directory should contain `selected_prefill_interpolation/` and `selected_decode_interpolation/` subdirectories with `raw_data.npz` files. This works seamlessly in Kubernetes where profile data is mounted via ConfigMap or PersistentVolume.
 
-To generate profiling data for your own model/hardware configuration, run the profiler (see [SLA-driven profiling documentation](../../../../docs/pages/components/profiler/profiler-guide.md) for details):
+To generate profiling data for your own model/hardware configuration, run the profiler (see [SLA-driven profiling documentation](../../../../docs/components/profiler/profiler-guide.md) for details):
 
 ```bash
 python components/src/dynamo/profiler/profile_sla.py \

components/src/dynamo/planner/README.md

@@ -42,5 +42,5 @@ When both modes are enabled, throughput-based scaling provides a lower bound on
 
 ## Documentation
 
-- **User docs**: [Planner Guide](../../../../docs/pages/components/planner/planner-guide.md) (deployment, configuration, examples)
-- **Design docs**: [Planner Design](../../../../docs/pages/design-docs/planner-design.md) (architecture, algorithms)
+- **User docs**: [Planner Guide](../../../../docs/components/planner/planner-guide.md) (deployment, configuration, examples)
+- **Design docs**: [Planner Design](../../../../docs/design-docs/planner-design.md) (architecture, algorithms)

components/src/dynamo/planner/utils/perf_interpolation.py

@@ -29,7 +29,7 @@ logger = logging.getLogger(__name__)
 
 MISSING_PROFILING_DATA_ERROR_MESSAGE = (
     "SLA-Planner requires pre-deployment profiling results to run.\n"
-    "Please follow /docs/pages/components/profiler/profiler-guide.md to run the profiling first,\n"
+    "Please follow /docs/components/profiler/profiler-guide.md to run the profiling first,\n"
     "and make sure the profiling results are present in --profile-results-dir."
 )
 

components/src/dynamo/profiler/README.md

@@ -6,8 +6,8 @@ SPDX-License-Identifier: Apache-2.0
 
 # Profiler
 
-Documentation for the Dynamo Profiler has moved to [docs/pages/components/profiler/](../../../../docs/pages/components/profiler/README.md).
+Documentation for the Dynamo Profiler has moved to [docs/components/profiler/](../../../../docs/components/profiler/README.md).
 
-- [Profiler Overview](../../../../docs/pages/components/profiler/README.md)
-- [Profiler Guide](../../../../docs/pages/components/profiler/profiler-guide.md)
-- [Profiler Examples](../../../../docs/pages/components/profiler/profiler-examples.md)
+- [Profiler Overview](../../../../docs/components/profiler/README.md)
+- [Profiler Guide](../../../../docs/components/profiler/profiler-guide.md)
+- [Profiler Examples](../../../../docs/components/profiler/profiler-examples.md)

components/src/dynamo/profiler/webui/utils.py

@@ -620,7 +620,7 @@ def create_gradio_interface(
 
             > 📝 **Note:** The dotted red line in the prefill and decode charts are default TTFT and ITL SLAs if not specified.
 
-            > ⚠️ **Warning:** The TTFT values here represent the ideal case when requests arrive uniformly, minimizing queueing. Real-world TTFT may be higher than profiling results. To mitigate the issue, planner uses [correction factors](https://github.com/ai-dynamo/dynamo/blob/main/docs/pages/design-docs/planner-design.md#step-2-correction-factor-calculation) to adjust dynamically at runtime.
+            > ⚠️ **Warning:** The TTFT values here represent the ideal case when requests arrive uniformly, minimizing queueing. Real-world TTFT may be higher than profiling results. To mitigate the issue, planner uses [correction factors](https://github.com/ai-dynamo/dynamo/blob/main/docs/design-docs/planner-design.md#step-2-correction-factor-calculation) to adjust dynamically at runtime.
 
             > 💡 **Tip:** Use the GPU cost checkbox and input in the charts section to convert GPU hours to cost.
             """

components/src/dynamo/router/README.md

@@ -3,7 +3,7 @@
 
 # Standalone Router
 
-A backend-agnostic standalone KV-aware router service for Dynamo deployments. For details on how KV-aware routing works, see the [Router Guide](/docs/pages/components/router/router-guide.md).
+A backend-agnostic standalone KV-aware router service for Dynamo deployments. For details on how KV-aware routing works, see the [Router Guide](/docs/components/router/router-guide.md).
 
 ## Overview
 
@@ -29,7 +29,7 @@ python -m dynamo.router \
 - `--endpoint`: Full endpoint path for workers in the format `namespace.component.endpoint` (e.g., `dynamo.prefill.generate`)
 
 **Router Configuration:**
-All router options use the `--router-*` prefix (e.g., `--router-block-size`, `--router-kv-overlap-score-weight`, `--router-temperature`, `--router-kv-events` / `--no-router-kv-events`, `--router-replica-sync`, `--router-snapshot-threshold`, `--router-reset-states`, `--router-track-active-blocks` / `--no-router-track-active-blocks`). Legacy names without the prefix (e.g., `--block-size`, `--kv-events`) are still accepted but deprecated. For detailed descriptions, see the [Router Guide](/docs/pages/components/router/router-guide.md).
+All router options use the `--router-*` prefix (e.g., `--router-block-size`, `--router-kv-overlap-score-weight`, `--router-temperature`, `--router-kv-events` / `--no-router-kv-events`, `--router-replica-sync`, `--router-snapshot-threshold`, `--router-reset-states`, `--router-track-active-blocks` / `--no-router-track-active-blocks`). Legacy names without the prefix (e.g., `--block-size`, `--kv-events`) are still accepted but deprecated. For detailed descriptions, see the [Router Guide](/docs/components/router/router-guide.md).
 
 ## Architecture
 
@@ -43,7 +43,7 @@ Clients call the `generate` endpoint to stream completions, or call `best_worker
 ## Example: Manual Disaggregated Serving (Alternative Setup)
 
 > [!Note]
-> **This is an alternative advanced setup.** The recommended approach for disaggregated serving is to use the frontend's automatic prefill routing, which activates when you register workers with `ModelType.Prefill`. See the [Router Guide](/docs/pages/components/router/router-guide.md#disaggregated-serving) for the default setup.
+> **This is an alternative advanced setup.** The recommended approach for disaggregated serving is to use the frontend's automatic prefill routing, which activates when you register workers with `ModelType.Prefill`. See the [Router Guide](/docs/components/router/router-guide.md#disaggregated-serving) for the default setup.
 >
 > Use this manual setup if you need explicit control over prefill routing configuration or want to manage prefill and decode routers separately.
 
@@ -103,7 +103,7 @@ See [`components/src/dynamo/vllm/handlers.py`](../vllm/handlers.py) for a refere
 
 ## See Also
 
-- [Router Guide](/docs/pages/components/router/router-guide.md) - Configuration and tuning for KV-aware routing
-- [Router Design](/docs/pages/design-docs/router-design.md) - Architecture details and event transport modes
+- [Router Guide](/docs/components/router/router-guide.md) - Configuration and tuning for KV-aware routing
+- [Router Design](/docs/design-docs/router-design.md) - Architecture details and event transport modes
 - [Frontend Router](../frontend/README.md) - Main HTTP frontend with integrated routing
 - [Router Benchmarking](/benchmarks/router/README.md) - Performance testing and tuning

components/src/dynamo/sglang/CLAUDE.md

@@ -230,7 +230,7 @@ text-to-video-diffusion.sh  # 1-2 GPUs - Text-to-video (Wan2.1)
   survives parent kill. Always check `nvidia-smi` after teardown.
 
 For troubleshooting (CuDNN, config.json errors, OOM, disagg connectivity), see
-`docs/pages/backends/sglang/sglang-examples.md#troubleshooting`.
+`docs/backends/sglang/sglang-examples.md#troubleshooting`.
 
 ## Adding a New Worker Type
 
@@ -268,7 +268,7 @@ Checklist for adding a new worker (e.g., a new modality or serving mode):
   pass engine=None. Guard any engine access in shared base class code.
 - **Rebuild after Rust changes**: If changing registration (register.py interacts with Rust
   bindings), rebuild: `cd lib/bindings/python && maturin develop --uv && cd <root> && uv pip install -e .`
-- **Troubleshooting**: See `docs/pages/backends/sglang/sglang-examples.md#troubleshooting`
+- **Troubleshooting**: See `docs/backends/sglang/sglang-examples.md#troubleshooting`
   for CuDNN, config.json, OOM, and disagg connectivity issues.
 
 ## File Index

components/src/dynamo/trtllm/engines/diffusion_engine.py

@@ -12,7 +12,7 @@ HuggingFace Diffusers model), eliminating the need for a --model-type flag.
 Requirements:
     - tensorrt_llm with visual_gen support (tensorrt_llm._torch.visual_gen).
       See: https://github.com/NVIDIA/TensorRT-LLM
-    - See docs/pages/backends/trtllm/README.md for setup instructions.
+    - See docs/backends/trtllm/README.md for setup instructions.
 
 Note on imports:
     tensorrt_llm._torch.visual_gen is imported lazily in initialize() because:

components/src/dynamo/trtllm/workers/video_diffusion_worker.py

@@ -133,7 +133,7 @@ async def init_video_diffusion_worker(
         raise RuntimeError(
             "ModelType.Videos not available in dynamo-runtime. "
             "Video diffusion requires a compatible dynamo-runtime version. "
-            "See docs/pages/backends/trtllm/README.md for setup instructions."
+            "See docs/backends/trtllm/README.md for setup instructions."
         )
     model_type = ModelType.Videos