reorg

Author: lmxyy

docs/source/conf.py

@@ -48,9 +48,7 @@
 # -- Options for HTML output -------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#options-for-html-output
 
-# html_permalinks_icon = "<span>#</span>"
-# html_theme = "sphinxawesome_theme"
-html_theme = "furo"
+html_theme = "sphinx_book_theme"
 html_static_path = ["_static"]
 
 napoleon_google_docstring = False

docs/source/index.rst

@@ -19,16 +19,17 @@ Check out `DeepCompressor <deepcompressor_repo>`_ for the quantization library.
 
 .. toctree::
     :maxdepth: 1
-    :caption: Advanced Features
-
-    advanced_features/lora.rst
-    advanced_features/kontext.rst
-    advanced_features/controlnet.rst
-    advanced_features/qencoder.rst
-    advanced_features/offload.rst
-    advanced_features/attention.rst
-    advanced_features/fbcache.rst
-    advanced_features/pulid.rst
+    :caption: Usage Tutorials
+
+    usage/basic_usage.rst
+    usage/lora.rst
+    usage/kontext.rst
+    usage/controlnet.rst
+    usage/qencoder.rst
+    usage/offload.rst
+    usage/attention.rst
+    usage/fbcache.rst
+    usage/pulid.rst
 
 .. toctree::
     :maxdepth: 1

…s/source/advanced_features/attention.rst docs/source/usage/attention.rstdocs/source/advanced_features/attention.rst renamed to docs/source/usage/attention.rst docs/source/advanced_features/attention.rst renamed to docs/source/usage/attention.rst

@@ -1,5 +1,5 @@
-Basic Usage Tutorial
-====================
+Basic Usage
+===========
 
 The following is a minimal script for running 4-bit `FLUX.1 <flux_repo_>`_ using Nunchaku.
 Nunchaku provides the same API as `Diffusers <diffusers_repo_>`_, so you can use it in a familiar way.
@@ -20,11 +20,16 @@ Nunchaku provides the same API as `Diffusers <diffusers_repo_>`_, so you can use
          :caption: Running FLUX.1-dev on Turing GPUs (`examples/flux.1-dev-turing.py <https://github.com/mit-han-lab/nunchaku/blob/main/examples/flux.1-dev-turing.py>`__)
          :linenos:
 
-The key difference when using Nunchaku is replacing the standard ``FluxTransformer2dModel`` with ``NunchakuFluxTransformer2dModel``. The ``NunchakuFluxTransformer2dModel.from_pretrained`` method loads quantized models and accepts either Hugging Face remote file paths or local file paths.
+The key difference when using Nunchaku is replacing the standard ``FluxTransformer2dModel`` 
+with :class:`~nunchaku.models.transformers.NunchakuFluxTransformer2dModel`.
+The :meth:`~nunchaku.models.transformers.NunchakuFluxTransformer2dModel.from_pretrained` 
+method loads quantized models and accepts either Hugging Face remote file paths or local file paths.
 
 .. note::
 
-   The ``get_precision()`` function automatically detects whether your GPU supports INT4 or FP4 quantization. Use FP4 models for Blackwell GPUs (RTX 50-series) and INT4 models for other architectures.
+   The :func:`~nunchaku.utils.get_precision` 
+   function automatically detects whether your GPU supports INT4 or FP4 quantization. 
+   Use FP4 models for Blackwell GPUs (RTX 50-series) and INT4 models for other architectures.
 
 .. note::