Sync from GitHub

.github/workflows/publish-to-hf-spaces.yml

@@ -0,0 +1,39 @@
+name: Publish to HF Spaces
+
+on:
+  push:
+    branches:
+      - main
+  workflow_dispatch:
+
+jobs:
+  publish:
+    runs-on: ubuntu-latest
+    env:
+      HF_REPO: diffusers/optimized-diffusers-code
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v3
+
+      - name: Install HF CLI
+        run: |
+          python -m pip install --upgrade pip
+          pip install huggingface_hub
+
+      - name: Log in to Hugging Face
+        run: |
+          huggingface-cli login --token "${{ secrets.HF_TOKEN }}"
+
+      - name: Create (or reuse) the Spaces repo
+        run: |
+          huggingface-cli repo create ${{env.HF_REPO}} \
+            --type space --space_sdk gradio  \
+            --exist-ok
+
+      - name: Push current repo to HF Spaces
+        run: |
+          huggingface-cli upload ${{env.HF_REPO}} \
+            . . \
+            --repo-type space \
+            --commit-message "Sync from GitHub" \
+            --token "${{ secrets.HF_TOKEN }}"

.gitignore

@@ -0,0 +1,195 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+__pycache__/
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# UV
+#   Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#uv.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/latest/usage/project/#working-with-version-control
+.pdm.toml
+.pdm-python
+.pdm-build/
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+
+# Abstra
+# Abstra is an AI-powered process automation framework.
+# Ignore directories containing user credentials, local state, and settings.
+# Learn more at https://abstra.io/docs
+.abstra/
+
+# Visual Studio Code
+#  Visual Studio Code specific template is maintained in a separate VisualStudioCode.gitignore 
+#  that can be found at https://github.com/github/gitignore/blob/main/Global/VisualStudioCode.gitignore
+#  and can be added to the global gitignore or merged into this file. However, if you prefer, 
+#  you could uncomment the following to ignore the enitre vscode folder
+# .vscode/
+
+# Ruff stuff:
+.ruff_cache/
+
+# PyPI configuration file
+.pypirc
+
+# Cursor
+#  Cursor is an AI-powered code editor. `.cursorignore` specifies files/directories to
+#  exclude from AI features like autocomplete and code analysis. Recommended for sensitive data
+#  refer to https://docs.cursor.com/context/ignore-files
+.cursorignore
+.cursorindexingignore

LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

README.md

@@ -1,12 +1,353 @@
----
-title: Optimized Diffusers Code
-emoji: 📈
-colorFrom: purple
-colorTo: purple
-sdk: gradio
-sdk_version: 5.34.2
-app_file: app.py
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# auto-diffusers-docs
+
+Still a WIP. Use an LLM to generate reasonable code snippets in a hardware-aware manner for Diffusers.
+
+### Motivation
+
+Within the Diffusers, we support a bunch of optimization techniques (refer [here](https://huggingface.co/docs/diffusers/main/en/optimization/memory), [here](https://huggingface.co/docs/diffusers/main/en/optimization/cache), and [here](https://huggingface.co/docs/diffusers/main/en/optimization/fp16)). However, it can be
+daunting for our users to determine when to use what. Hence, this repository tries to take a stab
+at using an LLM to generate reasonable code snippets for a given pipeline checkpoint that respects
+user hardware configuration.
+
+## Getting started
+
+Install the requirements from `requirements.txt`.
+
+Configure `GOOGLE_API_KEY` in the environment: `export GOOGLE_API_KEY=...`.
+
+Then run:
+
+```bash
+python e2e_example.py 
+```
+
+By default, the `e2e_example.py` script uses Flux.1-Dev, but this can be configured through the `--ckpt_id` argument.
+
+Full usage:
+
+```sh
+usage: e2e_example.py [-h] [--ckpt_id CKPT_ID] [--gemini_model GEMINI_MODEL] [--variant VARIANT] [--enable_lossy]
+
+options:
+  -h, --help            show this help message and exit
+  --ckpt_id CKPT_ID     Can be a repo id from the Hub or a local path where the checkpoint is stored.
+  --gemini_model GEMINI_MODEL
+                        Gemini model to use. Choose from https://ai.google.dev/gemini-api/docs/models.
+  --variant VARIANT     If the `ckpt_id` has variants, supply this flag to estimate compute. Example: 'fp16'.
+  --enable_lossy        When enabled, the code will include snippets for enabling quantization.
+```
+
+## Example outputs
+
+<details>
+<summary>python e2e_example.py (ran on an H100)</summary>
+
+````sh
+System RAM: 1999.99 GB
+RAM Category: large
+
+GPU VRAM: 79.65 GB
+VRAM Category: large
+current_generate_prompt='\npipeline_loading_memory_GB: 31.424\navailable_system_ram_GB: 1999.9855346679688\navailable_gpu_vram_GB: 79.6474609375\nenable_lossy_outputs: False\nenable_torch_compile: True\n'
+Sending request to Gemini...
+```python
+from diffusers import DiffusionPipeline
+import torch
+
+# User-provided information:
+# pipeline_loading_memory_GB: 31.424
+# available_system_ram_GB: 1999.9855346679688 (Large RAM)
+# available_gpu_vram_GB: 79.6474609375 (Large VRAM)
+# enable_lossy_outputs: False
+# enable_torch_compile: True
+
+# --- Configuration based on user needs and system capabilities ---
+
+# Placeholder for the actual checkpoint ID
+# Please replace this with your desired model checkpoint ID.
+CKPT_ID = "black-forest-labs/FLUX.1-dev" 
+
+# Determine dtype. bfloat16 is generally recommended for performance on compatible GPUs.
+# Ensure your GPU supports bfloat16 for optimal performance.
+dtype = torch.bfloat16
+
+# 1. Pipeline Loading and Device Placement:
+# Available VRAM (79.64 GB) is significantly greater than the pipeline's loading memory (31.42 GB).
+# Therefore, the entire pipeline can comfortably fit and run on the GPU.
+print(f"Loading pipeline '{CKPT_ID}' with {dtype} precision...")
+pipe = DiffusionPipeline.from_pretrained(CKPT_ID, torch_dtype=dtype)
+
+print("Moving pipeline to CUDA (GPU) as VRAM is sufficient...")
+pipe = pipe.to("cuda")
+
+# 2. Quantization:
+# User specified `enable_lossy_outputs: False`, so no quantization is applied.
+print("Quantization is NOT applied as per user's preference for lossless outputs.")
+
+# 3. Torch Compile:
+# User specified `enable_torch_compile: True`.
+# Since no offloading was applied (the entire model is on GPU), we can use `fullgraph=True`
+# for potentially greater performance benefits.
+print("Applying torch.compile() to the transformer for accelerated inference...")
+# The transformer is typically the most compute-intensive part of the diffusion pipeline.
+# Compiling it can lead to significant speedups.
+pipe.transformer.compile(fullgraph=True)
+
+# --- Inference ---
+print("Starting inference...")
+prompt = "photo of a dog sitting beside a river, high quality, 4k"
+image = pipe(prompt).images[0]
+
+print("Inference completed. Displaying image.")
+# Save or display the image
+image.save("generated_image.png")
+print("Image saved as generated_image.png")
+
+# You can also display the image directly if running in an environment that supports it
+# image.show()
+```
+````
+<br>
+</details>
+<br>
+<details>
+<summary>python e2e_example.py --enable_lossy</summary>
+
+````sh
+System RAM: 1999.99 GB
+RAM Category: large
+
+GPU VRAM: 79.65 GB
+VRAM Category: large
+current_generate_prompt='\npipeline_loading_memory_GB: 31.424\navailable_system_ram_GB: 1999.9855346679688\navailable_gpu_vram_GB: 79.6474609375\nenable_lossy_outputs: True\nenable_torch_compile: True\n'
+Sending request to Gemini...
+```python
+import torch
+from diffusers import DiffusionPipeline
+from diffusers.quantizers import PipelineQuantizationConfig
+import os
+
+# --- User-provided information and derived constants ---
+# Checkpoint ID (assuming a placeholder since it was not provided in the user input)
+# Using the example CKPT_ID from the problem description
+CKPT_ID = "black-forest-labs/FLUX.1-dev"
+
+# Derived from available_gpu_vram_GB (79.64 GB) and pipeline_loading_memory_GB (31.424 GB)
+# VRAM is ample to load the entire pipeline
+use_cuda_direct_load = True 
+
+# Derived from enable_lossy_outputs (True)
+enable_quantization = True
+
+# Derived from enable_torch_compile (True)
+enable_torch_compile = True
+
+# --- Inference Code ---
+
+print(f"Loading pipeline: {CKPT_ID}")
+
+# 1. Quantization Configuration (since enable_lossy_outputs is True)
+quant_config = None
+if enable_quantization:
+    # Default to bitsandbytes 4-bit as per guidance
+    print("Enabling bitsandbytes 4-bit quantization for 'transformer' component.")
+    quant_config = PipelineQuantizationConfig(
+        quant_backend="bitsandbytes_4bit", 
+        quant_kwargs={"load_in_4bit": True, "bnb_4bit_compute_dtype": torch.bfloat16, "bnb_4bit_quant_type": "nf4"},
+        # For FLUX.1-dev, the main generative component is typically 'transformer'.
+        # For other pipelines, you might include 'unet', 'text_encoder', 'text_encoder_2', etc.
+        components_to_quantize=["transformer"] 
+    )
+
+# 2. Load the Diffusion Pipeline
+# Use bfloat16 for better performance and modern GPU compatibility
+pipe = DiffusionPipeline.from_pretrained(
+    CKPT_ID, 
+    torch_dtype=torch.bfloat16,
+    quantization_config=quant_config if enable_quantization else None
+)
+
+# 3. Move Pipeline to GPU (since VRAM is ample)
+if use_cuda_direct_load:
+    print("Moving the entire pipeline to CUDA (GPU).")
+    pipe = pipe.to("cuda")
+
+# 4. Apply torch.compile() (since enable_torch_compile is True)
+if enable_torch_compile:
+    print("Applying torch.compile() for speedup.")
+    # This setting is beneficial when bitsandbytes is used
+    torch._dynamo.config.capture_dynamic_output_shape_ops = True 
+    
+    # Since no offloading is applied (model fits fully in VRAM), use fullgraph=True
+    # The primary component for compilation in FLUX.1-dev is 'transformer'
+    print("Compiling pipe.transformer with fullgraph=True.")
+    pipe.transformer = torch.compile(pipe.transformer, fullgraph=True)
+
+# 5. Perform Inference
+print("Starting image generation...")
+prompt = "photo of a dog sitting beside a river"
+num_inference_steps = 28 # A reasonable number of steps for good quality
+
+# Ensure all inputs are on the correct device for inference after compilation
+with torch.no_grad():
+    image = pipe(prompt, num_inference_steps=num_inference_steps).images[0]
+
+print("Image generation complete.")
+# Save or display the image
+output_path = "generated_image.png"
+image.save(output_path)
+print(f"Image saved to {output_path}")
+
+```
+````
+
+</details>
+<br>
+When invoked from an RTX 4090, it outputs:
+
+<details>
+<summary>Expand</summary>
+
+````sh
+System RAM: 125.54 GB
+RAM Category: large
+
+GPU VRAM: 23.99 GB
+VRAM Category: medium
+current_generate_prompt='\npipeline_loading_memory_GB: 31.424\navailable_system_ram_GB: 125.54026794433594\navailable_gpu_vram_GB: 23.98828125\nenable_lossy_outputs: False\nenable_torch_compile: True\n'
+Sending request to Gemini...
+```python
+import torch
+from diffusers import DiffusionPipeline
+import os # For creating offload directories if needed, though not directly used in this solution
+
+# --- User-provided information (interpreted) ---
+# Checkpoint ID will be a placeholder as it's not provided by the user directly in the input.
+# pipeline_loading_memory_GB: 31.424 GB
+# available_system_ram_GB: 125.54 GB (Categorized as "large": > 40GB)
+# available_gpu_vram_GB: 23.98 GB (Categorized as "medium": > 8GB <= 24GB)
+# enable_lossy_outputs: False (User prefers no quantization)
+# enable_torch_compile: True (User wants to enable torch.compile)
+
+# --- Configuration ---
+# Placeholder for the actual checkpoint ID. Replace with the desired model ID.
+CKPT_ID = "black-forest-labs/FLUX.1-dev" # Example from Diffusers library.
+PROMPT = "photo of a dog sitting beside a river"
+
+print(f"--- Optimizing inference for CKPT_ID: {CKPT_ID} ---")
+print(f"Pipeline loading memory: {31.424} GB")
+print(f"Available System RAM: {125.54} GB (Large)")
+print(f"Available GPU VRAM: {23.98} GB (Medium)")
+print(f"Lossy outputs (quantization): {'Disabled' if not False else 'Enabled'}")
+print(f"Torch.compile: {'Enabled' if True else 'Disabled'}")
+print("-" * 50)
+
+# --- 1. Load the Diffusion Pipeline ---
+# Use bfloat16 for a good balance of memory and performance.
+print(f"Loading pipeline '{CKPT_ID}' with torch_dtype=torch.bfloat16...")
+pipe = DiffusionPipeline.from_pretrained(CKPT_ID, torch_dtype=torch.bfloat16)
+print("Pipeline loaded.")
+
+# --- 2. Apply Memory Optimizations ---
+# Analysis:
+# - Pipeline memory (31.424 GB) exceeds available GPU VRAM (23.98 GB).
+# - System RAM (125.54 GB) is large.
+# Strategy: Use `enable_model_cpu_offload()`. This moves model components to CPU when not
+# in use, swapping them to GPU on demand. This is ideal when VRAM is insufficient but system
+# RAM is abundant.
+
+print("Applying memory optimization: `pipe.enable_model_cpu_offload()`...")
+pipe.enable_model_cpu_offload()
+print("Model CPU offloading enabled. Components will dynamically move between CPU and GPU.")
+
+# --- 3. Apply Speed Optimizations (torch.compile) ---
+# Analysis:
+# - `enable_torch_compile` is True.
+# - Model offloading (`enable_model_cpu_offload`) is applied.
+# Strategy: Enable torch.compile with `recompile_limit` as offloading is used.
+# Do not use `fullgraph=True` when offloading is active.
+
+print("Applying speed optimization: `torch.compile()`...")
+torch._dynamo.config.recompile_limit = 1000 # Recommended when offloading is applied.
+# torch._dynamo.config.capture_dynamic_output_shape_ops = True # Only for bitsandbytes, not applicable here.
+
+# Compile the main computational component (e.g., transformer or unet).
+# FLUX models primarily use a transformer. For other models, it might be `pipe.unet`.
+if hasattr(pipe, "transformer"):
+    print("Compiling `pipe.transformer`...")
+    pipe.transformer.compile()
+elif hasattr(pipe, "unet"):
+    print("Compiling `pipe.unet`...")
+    pipe.unet.compile()
+else:
+    print("Warning: Neither `pipe.transformer` nor `pipe.unet` found for compilation. Skipping `torch.compile` for core component.")
+
+print("Speed optimizations applied.")
+
+# --- 4. Perform Inference ---
+print(f"Starting image generation for prompt: '{PROMPT}'")
+# The pipeline handles component swapping automatically due to `enable_model_cpu_offload()`.
+image = pipe(PROMPT).images[0]
+print("Image generation complete.")
+
+# --- 5. Save or Display the Result ---
+output_path = "generated_image.png"
+image.save(output_path)
+print(f"Generated image saved to '{output_path}'")
+
+print("\n--- Inference process finished successfully ---")
+```
+````
+</details>
+
+### More outputs
+
+<details>
+<summary>"Wan-AI/Wan2.1-T2V-14B-Diffusers" with lossy outputs enabled</summary>
+
+````sh
+System RAM: 125.54 GB
+RAM Category: large
+
+GPU VRAM: 23.99 GB
+VRAM Category: medium
+("current_generate_prompt='\\nckpt_id: "
+ 'Wan-AI/Wan2.1-T2V-14B-Diffusers\\npipeline_loading_memory_GB: '
+ '37.432\\navailable_system_ram_GB: '
+ '125.54026794433594\\navailable_gpu_vram_GB: '
+ '23.98828125\\nenable_lossy_outputs: True\\nis_fp8_supported: '
+ "True\\nenable_torch_compile: True\\n'")
+Sending request to Gemini...
+```python
+from diffusers import DiffusionPipeline
+from diffusers.quantizers import PipelineQuantizationConfig
+import torch
+
+ckpt_id = "Wan-AI/Wan2.1-T2V-14B-Diffusers"
+
+quant_config = PipelineQuantizationConfig(
+    quant_backend="torchao",
+    quant_kwargs={"quant_type": "float8dq_e4m3_row"},
+    components_to_quantize=["transformer"]
+)
+pipe = DiffusionPipeline.from_pretrained(ckpt_id, quantization_config=quant_config, torch_dtype=torch.bfloat16)
+
+# Apply model CPU offload due to VRAM constraints
+pipe.enable_model_cpu_offload()
+
+# torch.compile() configuration
+torch._dynamo.config.recompile_limit = 1000
+pipe.transformer.compile()
+# pipe.vae.decode = torch.compile(pipe.vae.decode) # Uncomment if you want to compile VAE decode as well
+
+prompt = "photo of a dog sitting beside a river"
+
+# Modify the pipe call arguments as needed.
+image = pipe(prompt).images[0]
+
+# You can save the image or perform further operations here
+# image.save("generated_image.png")
+```
+````
+</details>
+<small>Ran on an RTX 4090</small>

app.py

@@ -0,0 +1,128 @@
+import gradio as gr
+from utils.pipeline_utils import determine_pipe_loading_memory
+from utils.llm_utils import LLMCodeOptimizer
+from prompts import system_prompt, generate_prompt
+from utils.hardware_utils import categorize_ram, categorize_vram
+
+LLM_CACHE = {}
+
+
+def get_output_code(
+    repo_id,
+    gemini_model_to_use,
+    disable_bf16,
+    enable_lossy,
+    system_ram,
+    gpu_vram,
+    torch_compile_friendly,
+    fp8_friendly,
+):
+    loading_mem_out = determine_pipe_loading_memory(repo_id, None, disable_bf16)
+    load_memory = loading_mem_out["total_loading_memory_gb"]
+    ram_category = categorize_ram(system_ram)
+    vram_category = categorize_vram(gpu_vram)
+
+    print(f"RAM Category: {ram_category}")
+    print(f"VRAM Category: {vram_category}")
+
+    if gemini_model_to_use not in LLM_CACHE:
+        print(f"Initializing new LLM instance for: {gemini_model_to_use}")
+        # If not, create it and add it to the cache
+        LLM_CACHE[gemini_model_to_use] = LLMCodeOptimizer(model_name=gemini_model_to_use, system_prompt=system_prompt)
+
+    llm = LLM_CACHE[gemini_model_to_use]
+    current_generate_prompt = generate_prompt.format(
+        ckpt_id=repo_id,
+        pipeline_loading_memory=load_memory,
+        available_system_ram=system_ram,
+        available_gpu_vram=gpu_vram,
+        enable_lossy_outputs=enable_lossy,
+        is_fp8_supported=fp8_friendly,
+        enable_torch_compile=torch_compile_friendly,
+    )
+    generated_prompt = current_generate_prompt
+    llm_output = llm(current_generate_prompt)
+    return llm_output, generated_prompt
+
+
+# --- Gradio UI Definition ---
+with gr.Blocks(theme=gr.themes.Soft()) as demo:
+    gr.Markdown(
+        """
+        # 🧨 Generate Diffusers Inference code snippet tailored to your machine
+        Enter a Hugging Face Hub `repo_id` and your system specs to get started for inference.
+        This tool uses [Gemini](https://ai.google.dev/gemini-api/docs/models) to generate the code based on your settings. This is based on
+        [sayakpaul/auto-diffusers-docs](https://github.com/sayakpaul/auto-diffusers-docs/).
+        """
+    )
+
+    with gr.Row():
+        with gr.Column(scale=3):
+            repo_id = gr.Textbox(
+                label="Hugging Face Repo ID",
+                placeholder="e.g., black-forest-labs/FLUX.1-dev",
+                info="The model repository you want to analyze.",
+                value="black-forest-labs/FLUX.1-dev",
+            )
+            gemini_model_to_use = gr.Dropdown(
+                ["gemini-2.5-flash", "gemini-2.5-pro"],
+                value="gemini-2.5-flash",
+                label="Gemini Model",
+                info="Select the model to generate the analysis.",
+            )
+            with gr.Row():
+                system_ram = gr.Number(label="System RAM (GB)", value=20)
+                gpu_vram = gr.Number(label="GPU VRAM (GB)", value=8)
+
+            with gr.Row():
+                disable_bf16 = gr.Checkbox(
+                    label="Disable BF16 (Use FP32)",
+                    value=False,
+                    info="Calculate using 32-bit precision instead of 16-bit.",
+                )
+                enable_lossy = gr.Checkbox(
+                    label="Allow Lossy Quantization", value=False, info="Consider 8-bit/4-bit quantization."
+                )
+                torch_compile_friendly = gr.Checkbox(
+                    label="torch.compile() friendly", value=False, info="Model is compatible with torch.compile."
+                )
+                fp8_friendly = gr.Checkbox(
+                    label="fp8 friendly", value=False, info="Model and hardware support FP8 precision."
+                )
+
+        with gr.Column(scale=1):
+            submit_btn = gr.Button("Estimate Memory ☁", variant="primary", scale=1)
+
+    with gr.Accordion("Generated LLM Prompt (for debugging)", open=False):
+        prompt_output = gr.Textbox(label="Prompt", show_copy_button=True, lines=10, interactive=False)
+
+    gr.Markdown("---")
+    gr.Markdown("### Estimation Result")
+
+    output_markdown = gr.Markdown(label="LLM Output", value="*Your results will appear here...*")
+
+    gr.Markdown(
+        """
+        ---
+        > ⛔️ **Disclaimer:** Large Language Models (LLMs) can make mistakes. The information provided
+        > is an estimate and should be verified. Always test the model on your target hardware to confirm
+        > actual memory requirements.
+        """
+    )
+
+    # --- Event Handling ---
+    all_inputs = [
+        repo_id,
+        gemini_model_to_use,
+        disable_bf16,
+        enable_lossy,
+        system_ram,
+        gpu_vram,
+        torch_compile_friendly,
+        fp8_friendly,
+    ]
+    submit_btn.click(fn=get_output_code, inputs=all_inputs, outputs=[output_markdown, prompt_output])
+
+
+if __name__ == "__main__":
+    demo.launch()

e2e_example.py

@@ -0,0 +1,92 @@
+import argparse
+from utils.llm_utils import LLMCodeOptimizer
+from prompts import system_prompt, generate_prompt
+from utils.pipeline_utils import determine_pipe_loading_memory
+from utils.hardware_utils import (
+    categorize_vram,
+    categorize_ram,
+    get_gpu_vram_gb,
+    get_system_ram_gb,
+    is_compile_friendly_gpu,
+    is_fp8_friendly,
+)
+import torch
+from pprint import pprint
+
+
+def create_parser():
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--ckpt_id",
+        type=str,
+        default="black-forest-labs/FLUX.1-dev",
+        help="Can be a repo id from the Hub or a local path where the checkpoint is stored.",
+    )
+    parser.add_argument(
+        "--gemini_model",
+        type=str,
+        default="gemini-2.5-flash-preview-05-20",
+        help="Gemini model to use. Choose from https://ai.google.dev/gemini-api/docs/models.",
+    )
+    parser.add_argument(
+        "--variant",
+        type=str,
+        default=None,
+        help="If the `ckpt_id` has variants, supply this flag to estimate compute. Example: 'fp16'.",
+    )
+    parser.add_argument(
+        "--disable_bf16",
+        action="store_true",
+        help="When enabled the load memory is affected. Prefer not enabling this flag.",
+    )
+    parser.add_argument(
+        "--enable_lossy",
+        action="store_true",
+        help="When enabled, the code will include snippets for enabling quantization.",
+    )
+    return parser
+
+
+def main(args):
+    if not torch.cuda.is_available():
+        raise ValueError("Not supported for non-CUDA devices for now.")
+    
+    loading_mem_out = determine_pipe_loading_memory(args.ckpt_id, args.variant, args.disable_bf16)
+    load_memory = loading_mem_out["total_loading_memory_gb"]
+    ram_gb = get_system_ram_gb()
+    ram_category = categorize_ram(ram_gb)
+    if ram_gb is not None:
+        print(f"\nSystem RAM: {ram_gb:.2f} GB")
+        print(f"RAM Category: {ram_category}")
+    else:
+        print("\nCould not determine System RAM.")
+
+    vram_gb = get_gpu_vram_gb()
+    vram_category = categorize_vram(vram_gb)
+    if vram_gb is not None:
+        print(f"\nGPU VRAM: {vram_gb:.2f} GB")
+        print(f"VRAM Category: {vram_category}")
+    else:
+        print("\nGPU VRAM check complete.")
+
+    is_compile_friendly = is_compile_friendly_gpu()
+    is_fp8_compatible = is_fp8_friendly()
+
+    llm = LLMCodeOptimizer(model_name=args.gemini_model, system_prompt=system_prompt)
+    current_generate_prompt = generate_prompt.format(
+        ckpt_id=args.ckpt_id,
+        pipeline_loading_memory=load_memory,
+        available_system_ram=ram_gb,
+        available_gpu_vram=vram_gb,
+        enable_lossy_outputs=args.enable_lossy,
+        is_fp8_supported=is_fp8_compatible,
+        enable_torch_compile=is_compile_friendly,
+    )
+    pprint(f"{current_generate_prompt=}")
+    print(llm(current_generate_prompt))
+
+
+if __name__ == "__main__":
+    parser = create_parser()
+    args = parser.parse_args()
+    main(args)

prompts.py

@@ -0,0 +1,183 @@
+system_prompt = """
+Consider yourself an expert at optimizing inference code for diffusion-based image and video generation models.
+For this project, you will be working with the Diffusers library. The library is built on top of PyTorch. Therefore,
+it's essential for you to exercise your PyTorch knowledge.
+
+Below is the simplest example of how a diffusion pipeline is usually used in Diffusers:
+
+```py
+from diffusers import DiffusionPipeline
+import torch
+
+ckpt_id = "black-forest-labs/FLUX.1-dev"
+pipe = DiffusionPipeline.from_pretrained(ckpt_id, torch_dtype=torch.bfloat16).to("cuda")
+image = pipe("photo of a dog sitting beside a river").images[0]
+```
+
+Your task will be to output a reasonable inference code in Python from user-supplied information about their
+needs. More specifically, you will be provided with the following information (in no particular order):
+
+* `ckpt_id` of the diffusion pipeline
+* Loading memory of a diffusion pipeline in GB
+* Available system RAM in GB
+* Available GPU VRAM in GB
+* If the user can afford to have lossy outputs (the likes of quantization)
+* If FP8 is supported
+* If the available GPU supports the latest `torch.compile()` knobs
+
+There are three categories of system RAM, broadly:
+
+* "small": <= 20GB
+* "medium": > 20GB <= 40GB
+* "large": > 40GB
+
+Similarly, there are three categories of VRAM, broadly:
+
+* "small": <= 8GB
+* "medium": > 8GB <= 24GB
+* "large": > 24GB
+
+Here is a high-level overview of what optimizations to apply for typical use cases.
+
+* Small VRAM, small system RAM
+
+Depending on the loading memory of the underlying diffusion pipeline, if the available VRAM and system RAM
+are both small, you apply a technique offloading called group offloading with disk serialization/deserialization
+support.
+
+Consider the code has an underlying component called `pipe` which has all the components needed
+to perform inference. So, the code for realizing the above solution would look something
+like so:
+
+```py
+from transformers import from transformers import PreTrainedModel
+# other imports go here.
+...
+
+onload_device = torch.device("cuda")
+pipe = DiffusionPipeline.from_pretrained(CKPT_ID, torch_dtype=torch.bfloat16)
+
+offload_dir = "DIRECTORY" # change me
+for name, module in pipe.components.items():
+    if hasattr(component, "_supports_group_offloading") and component._supports_group_offloading:
+        module.enable_group_offload(
+            onload_device=onload_device, 
+            offload_type="leaf_level", 
+            use_stream=True, 
+            offload_to_disk_path=f"{offload_dir}/{name}"
+        )
+    elif isinstance(component, PreTrainedModel):
+        apply_group_offloading(
+            module, 
+            onload_device=onload_device, 
+            offload_type="leaf_level",  
+            use_stream=True,
+            offload_to_disk_path=f"{offload_dir}/{name}"
+    )
+    
+# Inference goes here.
+...
+```
+
+* Small VRAM, medium system RAM
+
+Here, we can make use of model offloading:
+
+```py
+# other imports go here.
+...
+
+pipe = DiffusionPipeline.from_pretrained(CKPT_ID, torch_dtype=torch.bfloat16)
+pipe.enable_model_cpu_offload()
+
+# Inference goes here.
+...
+```
+
+* Large VRAM, large system RAM
+
+In this case, the `pipe` could directly be placed on CUDA if and only iff the loading memory requirements are
+satisfied by the available VRAM:
+
+```py
+pipe = pipe.to("cuda")
+```
+
+## Guidance on using quantization
+
+If the user specifies to use quantization, then you should default to using bitsandbytes 4bit. The code here
+would look like so:
+
+```py
+from diffusers.quantizers import PipelineQuantizationConfig
+# other imports go here.
+...
+
+quant_config = PipelineQuantizationConfig(
+    quant_backend="bitsandbytes_4bit", 
+    quant_kwargs={"load_in_4bit": True, "bnb_4bit_compute_dtype": torch.bfloat16, "bnb_4bit_quant_type": "nf4"}
+    components_to_quantize=["transformer"] # Can add a heavy text encoder here too.
+)
+pipe = DiffusionPipeline.from_pretrained(CKPT_ID, quantization_config=quant_config, torch_dtype=torch.bfloat16)
+
+# Rest of the code goes here.
+...
+```
+
+If there's support for performing FP8 computation, then we should use `torchao`:
+
+```py
+from diffusers.quantizers import PipelineQuantizationConfig
+# other imports go here.
+...
+
+quant_config = PipelineQuantizationConfig(
+    quant_backend="torchao", 
+    quant_kwargs={"quant_type": "float8dq_e4m3_row"}
+    components_to_quantize=["transformer"]
+)
+pipe = DiffusionPipeline.from_pretrained(CKPT_ID, quantization_config=quant_config, torch_dtype=torch.bfloat16)
+
+# Rest of the code goes here.
+...
+```
+
+## Guidance on using `torch.compile()`
+
+If the user wants to additionally boost inference speed, then you should the following line of code just before
+inference:
+
+* Add the following when offloading was applied: `torch._dynamo.config.recompile_limit = 1000`.
+* ONLY, add the following when `bitsandbytes` was used for `quant_backend`: `torch._dynamo.config.capture_dynamic_output_shape_ops = True`.
+* Finally, add `pipe.transformer.compile()`.
+* Add `pipe.vae.decode = torch.compile(vae.decode)` as a comment.
+
+In case no offloading was applied, then the line should be:
+
+```py
+pipe.transformer.compile(fullgraph=True)
+```
+
+## Other guidelines
+
+* When the available VRAM > pipeline loading memory, you should suggest using `pipe = pipe.to("cuda")`.
+* If the user prefers not to use quantization and further reduce memory, then suggest using:
+`pipe.transformer.enable_layerwise_casting(storage_dtype=torch.float8_e4m3fn, compute_dtype=torch.bfloat16)`.
+* Do NOT add any extra imports or lines of code that will not be used. 
+* Do NOT try to be too creative about combining the optimization techniques laid out above.
+* Do NOT add extra arguments to the `pipe` call other than the `prompt`.
+* Add a comment before the `pipe` call, saying "Modify the pipe call arguments as needed."
+* Do NOT add any serialization step after the pipe call.
+
+Please think about these guidelines carefully before producing the outputs.
+"""
+
+generate_prompt = """
+ckpt_id: {ckpt_id}
+pipeline_loading_memory_GB: {pipeline_loading_memory}
+available_system_ram_GB: {available_system_ram}
+available_gpu_vram_GB: {available_gpu_vram}
+enable_lossy_outputs: {enable_lossy_outputs}
+is_fp8_supported: {is_fp8_supported}
+enable_torch_compile: {enable_torch_compile}
+"""

requirements.txt

@@ -0,0 +1,5 @@
+torch>=2.1
+safetensors>=0.5.3
+huggingface_hub>=0.30.2
+hf_xet
+google-generativeai>=0.8.5

utils/hardware_utils.py

@@ -0,0 +1,128 @@
+import subprocess
+import psutil
+import functools
+from torch._inductor.runtime.hints import DeviceProperties
+from torch._inductor.utils import get_gpu_type
+from typing import Union
+import torch
+
+
+@functools.cache
+def get_system_ram_gb():
+    """
+    Gets the total physical system RAM in Gigabytes.
+
+    Returns:
+        float: Total system RAM in GB, or None if it cannot be determined.
+    """
+    try:
+        # Get virtual memory details
+        virtual_memory = psutil.virtual_memory()
+        # Total physical memory in bytes
+        total_ram_bytes = virtual_memory.total
+        # Convert bytes to gigabytes (1 GB = 1024^3 bytes)
+        total_ram_gb = total_ram_bytes / (1024**3)
+        return total_ram_gb
+    except Exception as e:
+        print(f"Error getting system RAM: {e}")
+        return None
+
+
+@functools.cache
+def get_gpu_vram_gb():
+    """
+    Gets the total GPU VRAM in Gigabytes using the nvidia-smi command.
+    This function is intended for NVIDIA GPUs.
+
+    Returns:
+        float: Total GPU VRAM in GB, or None if it cannot be determined.
+    """
+    try:
+        # Execute the nvidia-smi command to get GPU memory info
+        # The command queries for the total memory and outputs it in MiB
+        result = subprocess.run(
+            ["nvidia-smi", "--query-gpu=memory.total", "--format=csv,noheader,nounits"],
+            capture_output=True,
+            text=True,
+            check=True,
+        )
+        # The output will be a string like "12288\n" for the first GPU
+        # We take the first line in case there are multiple GPUs
+        vram_mib = int(result.stdout.strip().split("\n")[0])
+        # Convert MiB to Gigabytes (1 GB = 1024 MiB)
+        vram_gb = vram_mib / 1024
+        return vram_gb
+    except FileNotFoundError:
+        # This error occurs if nvidia-smi is not installed or not in the PATH
+        print("INFO: 'nvidia-smi' command not found. Cannot determine GPU VRAM.")
+        print("      This is expected if you don't have an NVIDIA GPU or drivers installed.")
+        return None
+    except (subprocess.CalledProcessError, ValueError, IndexError) as e:
+        # Handles other potential errors like command failure or parsing issues
+        print(f"Error getting GPU VRAM: {e}")
+        return None
+
+
+def categorize_ram(ram_gb):
+    """
+    Categorizes RAM into 'small', 'medium', or 'large'.
+
+    Args:
+        ram_gb (float): The amount of RAM in GB.
+
+    Returns:
+        str: The category ('small', 'medium', 'large') or 'unknown'.
+    """
+    if ram_gb is None:
+        return "unknown"
+    if ram_gb <= 20:
+        return "small"
+    elif 20 < ram_gb <= 40:
+        return "medium"
+    else:  # ram_gb > 40
+        return "large"
+
+
+def categorize_vram(vram_gb):
+    """
+    Categorizes VRAM into 'small', 'medium', or 'large'.
+
+    Args:
+        vram_gb (float): The amount of VRAM in GB.
+
+    Returns:
+        str: The category ('small', 'medium', 'large') or 'not applicable/unknown'.
+    """
+    if vram_gb is None:
+        return "not applicable/unknown"
+    if vram_gb <= 8:
+        return "small"
+    elif 8 < vram_gb <= 24:
+        return "medium"
+    else:  # vram_gb > 24
+        return "large"
+
+
+@functools.cache
+def is_compile_friendly_gpu(index_or_device: Union[int, str, torch.device] = 0) -> bool:
+    """Hand-coded rules from experiments. Don't take seriously."""
+    if isinstance(index_or_device, torch.device):
+        device = index_or_device
+    elif isinstance(index_or_device, str):
+        device = torch.device(index_or_device)
+    else:
+        device = torch.device(get_gpu_type(), index_or_device)
+
+    prop = DeviceProperties.create(device)
+    return prop.major >= 8
+
+
+@functools.lru_cache()
+def is_sm_version(major: int, minor: int) -> bool:
+    """Check if the CUDA version is exactly major.minor"""
+    is_cuda = torch.cuda.is_available() and torch.version.cuda
+    return torch.cuda.get_device_capability() == (major, minor) if is_cuda else False
+
+
+def is_fp8_friendly():
+    return is_sm_version(8, 9)

utils/llm_utils.py

@@ -0,0 +1,26 @@
+import google.generativeai as genai
+import os
+
+# https://ai.google.dev/gemini-api/docs/models#gemini-2.5-flash-preview
+MODEL_NAME = "gemini-2.5-flash-preview-05-20"
+
+
+class LLMCodeOptimizer:
+    def __init__(self, model_name: str, system_prompt: str):
+        api_key = os.getenv("GOOGLE_API_KEY")
+        if api_key is None:
+            raise ValueError("Must provide an API key for Gemini through the `GOOGLE_API_KEY` env variable.")
+        genai.configure(api_key=api_key)
+        self.model = genai.GenerativeModel(model_name=model_name, system_instruction=system_prompt)
+
+    def __call__(self, generation_prompt):
+        try:
+            print("Sending request to Gemini...")
+            response = self.model.generate_content(generation_prompt)
+
+            return response.text
+
+        except Exception as e:
+            # Handle potential exceptions, such as invalid API keys,
+            # network issues, or content moderation errors.
+            return f"An error occurred: {e}"

utils/pipeline_utils.py

@@ -0,0 +1,194 @@
+import collections
+from pathlib import Path
+import functools
+import os
+import safetensors.torch
+from huggingface_hub import model_info, hf_hub_download
+import tempfile
+import torch
+import functools
+import os
+import requests
+import struct
+from huggingface_hub import hf_hub_url
+
+DTYPE_MAP = {"FP32": torch.float32, "FP16": torch.float16, "BF16": torch.bfloat16}
+
+
+# https://huggingface.co/docs/safetensors/v0.3.2/metadata_parsing#python
+def _parse_single_file(url):
+    print(f"{url=}")
+    token = os.getenv("HF_TOKEN")
+    headers = {"Range": "bytes=0-7", "Authorization": f"Bearer {token}"}
+    response = requests.get(url, headers=headers)
+    length_of_header = struct.unpack("<Q", response.content)[0]
+    headers = {"Range": f"bytes=8-{7 + length_of_header}", "Authorization": f"Bearer {token}"}
+    response = requests.get(url, headers=headers)
+    header = response.json()
+    return header
+
+
+def _get_dtype_from_safetensor_file(file_path):
+    """Inspects a safetensors file and returns the dtype of the first tensor.
+
+    If it's not a safetensors file and a URL instead, we query it.
+    """
+    if "https" in file_path:
+        metadata = _parse_single_file(file_path)
+        except_format_metadata_keys = sorted({k for k in metadata if k != "__metadata__"})
+        string_dtype = metadata[except_format_metadata_keys[0]]["dtype"]
+        return DTYPE_MAP[string_dtype]
+    try:
+        # load_file is simple and sufficient for this info-gathering purpose.
+        state_dict = safetensors.torch.load_file(file_path)
+        if not state_dict:
+            return "N/A (empty)"
+
+        # Get the dtype from the first tensor in the state dict
+        first_tensor = next(iter(state_dict.values()))
+        return first_tensor.dtype
+    except Exception as e:
+        print(f"Warning: Could not determine dtype from {file_path}. Error: {e}")
+        return "N/A (error)"
+
+
+def _process_components(component_files, file_accessor_fn, disable_bf16=False):
+    """
+    Generic function to process components, calculate size, and determine dtype.
+
+    Args:
+        component_files (dict): A dictionary mapping component names to lists of file objects.
+        file_accessor_fn (function): A function that takes a file object and returns
+                                     a tuple of (local_path_for_inspection, size_in_bytes, relative_filename).
+        disable_bf16 (bool): To disable using `torch.bfloat16`. Use it at your own risk.
+
+    Returns:
+        dict: A dictionary containing the total memory and detailed component info.
+    """
+    components_info = {}
+    total_size_bytes = 0
+
+    for name, files in component_files.items():
+        # Get dtype by inspecting the first file of the component
+        first_file = files[0]
+
+        # The accessor function handles how to get the path (download vs local)
+        # and its size and relative name.
+        inspection_path, _, _ = file_accessor_fn(first_file)
+        dtype = _get_dtype_from_safetensor_file(inspection_path)
+
+        component_size_bytes = 0
+        component_file_details = []
+        for f in files:
+            _, size_bytes, rel_filename = file_accessor_fn(f)
+            component_size_bytes += size_bytes
+            component_file_details.append({"filename": rel_filename, "size_mb": size_bytes / (1024**2)})
+
+        if dtype == torch.float32 and not disable_bf16:
+            print(
+                f"The `dtype` for component ({name}) is torch.float32. Since bf16 computation is not disabled "
+                "we will slash the total size of this component by 2."
+            )
+            total_size_bytes += component_size_bytes / 2
+        else:
+            total_size_bytes += component_size_bytes
+
+        components_info[name] = {
+            "size_gb": round(component_size_bytes / (1024**3), 3),
+            "dtype": dtype,
+            "files": sorted(component_file_details, key=lambda x: x["filename"]),
+        }
+
+    return {
+        "total_loading_memory_gb": round(total_size_bytes / (1024**3), 3),
+        "components": components_info,
+    }
+
+
+@functools.lru_cache()
+def _determine_memory_from_hub_ckpt(ckpt_id, variant=None, disable_bf16=False):
+    """
+    Determines memory and dtypes for a checkpoint on the Hugging Face Hub.
+    """
+    files_in_repo = model_info(ckpt_id, files_metadata=True, token=os.getenv("HF_TOKEN")).siblings
+    all_safetensors_siblings = [
+        s for s in files_in_repo if s.rfilename.endswith(".safetensors") and "/" in s.rfilename
+    ]
+    if variant:
+        all_safetensors_siblings = [f for f in all_safetensors_siblings if variant in f.rfilename]
+
+    component_files = collections.defaultdict(list)
+    for sibling in all_safetensors_siblings:
+        component_name = Path(sibling.rfilename).parent.name
+        component_files[component_name].append(sibling)
+
+    with tempfile.TemporaryDirectory() as temp_dir:
+
+        def hub_file_accessor(file_obj):
+            """Accessor for Hub files: downloads them and returns path/size."""
+            print(f"Querying '{file_obj.rfilename}' for inspection...")
+            url = hf_hub_url(ckpt_id, file_obj.rfilename)
+            return url, file_obj.size, file_obj.rfilename
+
+        # We only need to download one file per component for dtype inspection.
+        # To make this efficient, we create a specialized accessor for the processing loop
+        # that only downloads the *first* file encountered for a component.
+        downloaded_for_inspection = {}
+
+        def efficient_hub_accessor(file_obj):
+            component_name = Path(file_obj.rfilename).parent.name
+            if component_name not in downloaded_for_inspection:
+                path, _, _ = hub_file_accessor(file_obj)
+                downloaded_for_inspection[component_name] = path
+
+            inspection_path = downloaded_for_inspection[component_name]
+            return inspection_path, file_obj.size, file_obj.rfilename
+
+        return _process_components(component_files, efficient_hub_accessor, disable_bf16)
+
+
+@functools.lru_cache()
+def _determine_memory_from_local_ckpt(path: str, variant=None, disable_bf16=False):
+    """
+    Determines memory and dtypes for a local checkpoint.
+    """
+    ckpt_path = Path(path)
+    if not ckpt_path.is_dir():
+        return {"error": f"Checkpoint path '{path}' not found or is not a directory."}
+
+    all_safetensors_paths = list(ckpt_path.glob("**/*.safetensors"))
+    if variant:
+        all_safetensors_paths = [p for p in all_safetensors_paths if variant in p.name]
+
+    component_files = collections.defaultdict(list)
+    for file_path in all_safetensors_paths:
+        component_name = file_path.parent.name
+        component_files[component_name].append(file_path)
+
+    def local_file_accessor(file_path):
+        """Accessor for local files: just returns their path and size."""
+        return file_path, file_path.stat().st_size, str(file_path.relative_to(ckpt_path))
+
+    return _process_components(component_files, local_file_accessor, disable_bf16)
+
+
+def determine_pipe_loading_memory(ckpt_id: str, variant=None, disable_bf16=False):
+    """
+    Determines the memory and dtypes for a pipeline, whether it's local or on the Hub.
+    """
+    if os.path.isdir(ckpt_id):
+        return _determine_memory_from_local_ckpt(ckpt_id, variant, disable_bf16)
+    else:
+        return _determine_memory_from_hub_ckpt(ckpt_id, variant, disable_bf16)
+
+
+if __name__ == "__main__":
+    output = _determine_memory_from_hub_ckpt("Wan-AI/Wan2.1-T2V-14B-Diffusers")
+    total_size_gb = output["total_loading_memory_gb"]
+    safetensor_files = output["components"]
+    print(f"{total_size_gb=} GB")
+    print(f"{safetensor_files=}")
+    print("\n")
+    # total_size_gb, safetensor_files = _determine_memory_from_local_ckpt("LOCAL_DIR") # change me.
+    # print(f"{total_size_gb=} GB")
+    # print(f"{safetensor_files=}")