Remove replace_module_for_calibration

[dsikka]

SUMMARY:

• Remove prepare.py
• Remove replace_modules_for_calibration from examples
• Update context

TEST PLAN:

Running through all our examples

1. Accuracy: Qwen3-30B-A3B-NVFP4 - 99% recovery

|Tasks|Version|     Filter     |n-shot|  Metric   |   |Value |   |Stderr|
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|gsm8k|      3|flexible-extract|     5|exact_match|↑  |0.8848|±  |0.0088|
|     |       |strict-match    |     5|exact_match|↑  |0.8840|±  |0.0088|

2. Accuracy: Qwen3-VL-235B-A22B-Instruct-NVFP4

|Tasks|Version|     Filter     |n-shot|  Metric   |   |Value |   |Stderr|
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|gsm8k|      3|flexible-extract|     5|exact_match|↑  |0.9037|±  |0.0081|
|     |       |strict-match    |     5|exact_match|↑  |0.8916|±  |0.0086|

| Tasks  |Version|Filter|n-shot| Metric |   |Value |   |Stderr|
|--------|------:|------|-----:|--------|---|-----:|---|------|
|mmmu_val|      0|none  |     0|mmmu_acc|↑  |0.6178|±  |   N/A|

3. Accuracy: Qwen3-VL-30B-A3B-Instruct-AWQ-W8A16-mse-seq

Greater than 99% Recovery

|Tasks|Version|     Filter     |n-shot|  Metric   |   |Value |   |Stderr|
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|gsm8k|      3|flexible-extract|     5|exact_match|↑  |0.8817|±  |0.0089|
|     |       |strict-match    |     5|exact_match|↑  |0.8719|±  |0.0092|

Greater than 99% Recovery

| Tasks  |Version|Filter|n-shot| Metric |   |Value |   |Stderr|
|--------|------:|------|-----:|--------|---|-----:|---|------|
|mmmu_val|      0|none  |     0|mmmu_acc|↑  |0.5078|±  |   N/A|

4. Accuracy Qwen3-VL-235B-A22B-Instruct-FP8-DYNAMIC
   Similar accuracy as above

[github-actions]

👋 Hi! Thank you for contributing to llm-compressor. Please add the ready label when the PR is ready for review.

Note: This is required to complete the testing suite, please only add the label once the PR is code complete and local testing has been performed.

[gemini-code-assist]

Summary of Changes

Hello @dsikka, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request refines the model preparation and quantization examples by eliminating a specific module replacement for Qwen3VLMoeTextSparseMoeBlock. This change simplifies the replace_modules_for_calibration function's internal configuration and removes its invocation from relevant example scripts, indicating that this particular replacement is no longer necessary for current calibration workflows.

Highlights

• Module Replacement Logic Removal: The specific module replacement logic for Qwen3VLMoeTextSparseMoeBlock has been removed from the src/llmcompressor/modeling/prepare.py file. This includes removing its import and its entry from the internal replacements dictionary.
• Example Code Simplification: Calls to the replace_modules_for_calibration(model) function have been removed from two quantization examples: examples/quantization_w4a4_fp4/qwen3_vl_moe_w4a4_fp4.py and examples/quantization_w8a8_fp8/qwen3_vl_moe_fp8_example.py.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request correctly removes the usage of the deprecated replace_modules_for_calibration function for the Qwen3-VL-MoE model, which is a good cleanup. I've added one minor style suggestion. Additionally, this change makes some other code redundant which could be cleaned up in a follow-up: the replace function in src/llmcompressor/modeling/qwen3_vl_moe.py and the unused imports in the updated example files are now obsolete.

[kylesayrs]

Sweet, thanks