Reorganization of ToC

mtsokol · mtsokol · commit 52717ee16fca · 2026-03-27T16:48:34.000+01:00
Signed-off-by: Mateusz Sokół &lt;mat646@gmail.com&gt;
diff --git a/docs/README.md b/docs/README.md
@@ -22,7 +22,7 @@ vLLM TPU is now powered by `tpu-inference`, an expressive and powerful new hardw
 
 Although vLLM TPU’s new unified backend makes out-of-the-box high performance serving possible with any model supported in vLLM, the reality is that we're still in the process of implementing a few core components.
 
-For this reason, we’ve provided a **[Recommended Models and Features](recommended_models_features.md)** page detailing the models and features that are validated through unit, integration, and performance testing.
+For this reason, we’ve provided a **[Recommended Models and Features](basic_usage/recommended_models.md)** page detailing the models and features that are validated through unit, integration, and performance testing.
 
 ## Getting Started
 
@@ -44,8 +44,8 @@ Tested end-to-end guides for hosting specific models on specific TPU generations
 
 If you are interested in contributing to the project or want to learn more about the internals, check out our developer guides:
 
-- **[JAX Model Development](developer_guides/jax_model_development.md)**
-- **[Torch Model Development](developer_guides/torchax_model_development.md)**
+- **[JAX Model Development](advanced_usage/jax_model_development.md)**
+- **[Torch Model Development](advanced_usage/torchax_model_development.md)**
 
 ## Contribute
 
diff --git a/docs/advanced_usage/gce_example.md b/docs/advanced_usage/gce_example.md
@@ -0,0 +1,3 @@
+# Run & server your custom model on GCE
+
+...
diff --git a/docs/advanced_usage/jax_model_development.md b/docs/advanced_usage/jax_model_development.md
diff --git a/docs/advanced_usage/jax_overview.md b/docs/advanced_usage/jax_overview.md
@@ -0,0 +1,3 @@
+# A mental model how JAX-based vLLM model runs on TPU
+
+...
diff --git a/docs/advanced_usage/recommended_features.md b/docs/advanced_usage/recommended_features.md
@@ -0,0 +1,23 @@
+## Recommended Features
+
+This table shows the features currently tested for accuracy and performance.
+
+{{ read_csv('../support_matrices/feature_support_matrix.csv', keep_default_na=False) }}
+
+## Kernel Support
+
+This table shows the current kernel support status.
+
+{{ read_csv('../support_matrices/kernel_support_matrix.csv', keep_default_na=False) }}
+
+## Parallelism Support
+
+This table shows the current parallelism support status.
+
+{{ read_csv('../support_matrices/parallelism_support_matrix.csv', keep_default_na=False) }}
+
+## Quantization Support
+
+This table shows the current quantization support status.
+
+{{ read_csv('../support_matrices/quantization_support_matrix.csv', keep_default_na=False) }}
diff --git a/docs/advanced_usage/torch_overview.md b/docs/advanced_usage/torch_overview.md
@@ -0,0 +1,3 @@
+# How is torchax used in vLLM
+
+...
diff --git a/docs/advanced_usage/torchax_model_development.md b/docs/advanced_usage/torchax_model_development.md
@@ -1,4 +1,4 @@
-# How is torchax used in vLLM
+# E2E code example
 
 **Authors: Siyuan Liu, Hongmin Fan, Han Qi**
 
diff --git a/docs/basic_usage/overview.md b/docs/basic_usage/overview.md
diff --git a/docs/basic_usage/recommended_models.md b/docs/basic_usage/recommended_models.md
@@ -14,27 +14,3 @@ These tables show the models currently tested for accuracy and performance.
 ### Models
 
 {{ read_csv('../support_matrices/model_support_matrix.csv', keep_default_na=False) }}
-
-## Recommended Features
-
-This table shows the features currently tested for accuracy and performance.
-
-{{ read_csv('../support_matrices/feature_support_matrix.csv', keep_default_na=False) }}
-
-## Kernel Support
-
-This table shows the current kernel support status.
-
-{{ read_csv('../support_matrices/kernel_support_matrix.csv', keep_default_na=False) }}
-
-## Parallelism Support
-
-This table shows the current parallelism support status.
-
-{{ read_csv('../support_matrices/parallelism_support_matrix.csv', keep_default_na=False) }}
-
-## Quantization Support
-
-This table shows the current quantization support status.
-
-{{ read_csv('../support_matrices/quantization_support_matrix.csv', keep_default_na=False) }}
diff --git a/docs/contributors_guide.md b/docs/contributors_guide.md
@@ -0,0 +1,7 @@
+# How to contribute
+
+...
+
+## CoC
+
+...
diff --git a/docs/faq.md b/docs/faq.md
@@ -0,0 +1,3 @@
+# FAQ and Sharp Bits
+
+...
diff --git a/docs/mkdocs/stylesheets/extra.css b/docs/mkdocs/stylesheets/extra.css
@@ -0,0 +1,11 @@
+/* Light mode: darker section titles */
+body[data-md-color-scheme="default"] .md-nav__item--section > label.md-nav__link .md-ellipsis {
+  color: rgba(0, 0, 0, 0.7) !important;
+  font-weight: 700;
+}
+
+/* Dark mode: lighter gray section titles */
+body[data-md-color-scheme="slate"] .md-nav__item--section > label.md-nav__link .md-ellipsis {
+  color: rgba(255, 255, 255, 0.8) !important;
+  font-weight: 700;
+}
diff --git a/mkdocs.yml b/mkdocs.yml
@@ -59,16 +59,30 @@ nav:
   - Getting Started:
     - TPU Setup: getting_started/tpu_setup.md
     - Installation: getting_started/installation.md
-    - Quickstart: getting_started/quickstart.md
-    - 'Out-of-Tree Usage': getting_started/out-of-tree.md
-  - 'Recommended Models and Features': recommended_models_features.md
-  - Developer Guides:
-    - 'JAX Model Development': developer_guides/jax_model_development.md
-    - 'Torch Model Development': developer_guides/torchax_model_development.md
-  - 'Recipes (External)':
-    - 'v7x (Ironwood)': https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/ironwood/vLLM
-    - 'v6e (Trillium)': https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/trillium/vLLM
-  - 'Profiling' : profiling.md
+    - Serving model: getting_started/quickstart.md
+  - 'Basic Usage (I use HuggingFace model)':
+    - 'Overview & TPU': basic_usage/overview.md
+    - 'Recommended Models': basic_usage/recommended_models.md
+    - 'Deploying model on GCE':
+      - 'v7x setup (Ironwood)': https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/ironwood/vLLM
+      - 'v6e setup (Trillium)': https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/trillium/vLLM
+    - 'Code Examples':
+      - 'Multi Modal Inference': https://github.com/vllm-project/tpu-inference/blob/main/examples/multi_modal_inference.py
+      - 'Offline LoRA': https://github.com/vllm-project/tpu-inference/blob/main/examples/offline_lora_inference.py
+  - Advanced Usage (I need my custom model):
+    - 'JAX':
+      - 'How it works': advanced_usage/jax_overview.md
+      - 'Model Development': advanced_usage/jax_model_development.md
+    - 'Torch':
+      - 'How it works': advanced_usage/torch_overview.md
+      - 'Model Development': advanced_usage/torchax_model_development.md
+    - 'Your custom model on GCE': advanced_usage/gce_example.md
+    - 'Recommended Features': advanced_usage/recommended_features.md
+  - 'See Also':
+    - 'Profiling' : profiling.md
+    - 'FAQ': faq.md
+    - 'Contributors Guide': contributors_guide.md
+
 plugins:
   - search
   - same-dir
@@ -107,6 +121,9 @@ markdown_extensions:
   - mdx_math:
       enable_dollar_delimiter: true
 
+extra_css:
+  - mkdocs/stylesheets/extra.css
+
 extra_javascript:
   - https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS_HTML
 

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,3 @@`
	`1`	`+# Run & server your custom model on GCE`
	`2`	`+`
	`3`	`+...`
Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,3 @@`
	`1`	`+# A mental model how JAX-based vLLM model runs on TPU`
	`2`	`+`
	`3`	`+...`
Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,3 @@`
	`1`	`+# How is torchax used in vLLM`
	`2`	`+`
	`3`	`+...`
Original file line number	Diff line number	Diff line change
`@@ -1,4 +1,4 @@`
`1`		`-# How is torchax used in vLLM`
	`1`	`+# E2E code example`
`2`	`2`
`3`	`3`	`Authors: Siyuan Liu, Hongmin Fan, Han Qi`
`4`	`4`
-Original file line number
+Diff line change
@@ @@ -0,0 +1,7 @@ @@
 +# How to contribute
++
 +...
++
 +## CoC
++
 +...