[Bugfix] Skip loading extra parameters for modelopt Qwen3 MoE model

[noiji]

Essential Elements of an Effective PR Description Checklist

• The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
• The test plan, such as providing test command.
• The test results, such as pasting the results comparison before and after, or e2e results
• (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.

Purpose

Fixes #15784

Test Plan

add the following to the quantization_config field of config.json

"quant_method": "fp8",
"activation_scheme": "static",

and then serve with

vllm serve {modelopt_quantized_model}

Test Result

(AS-IS)

ERROR 06-13 14:37:58 [core.py:515]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/qwen3_moe.py", line 535, in load_weights
ERROR 06-13 14:37:58 [core.py:515]     return loader.load_weights(weights)
ERROR 06-13 14:37:58 [core.py:515]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 06-13 14:37:58 [core.py:515]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/utils.py", line 278, in load_weights
ERROR 06-13 14:37:58 [core.py:515]     autoloaded_weights = set(self._load_module("", self.module, weights))
ERROR 06-13 14:37:58 [core.py:515]                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 06-13 14:37:58 [core.py:515]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/utils.py", line 236, in _load_module
ERROR 06-13 14:37:58 [core.py:515]     yield from self._load_module(prefix,
ERROR 06-13 14:37:58 [core.py:515]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/utils.py", line 209, in _load_module
ERROR 06-13 14:37:58 [core.py:515]     loaded_params = module_load_weights(weights)
ERROR 06-13 14:37:58 [core.py:515]                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 06-13 14:37:58 [core.py:515]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/qwen3_moe.py", line 469, in load_weights
ERROR 06-13 14:37:58 [core.py:515]     param = params_dict[name]
ERROR 06-13 14:37:58 [core.py:515]             ~~~~~~~~~~~^^^^^^
ERROR 06-13 14:37:58 [core.py:515] KeyError: 'layers.40.self_attn.qkqkv_proj.k_scale'

(TO-BE)

• loads the model successfully

(Optional) Documentation Update

[gemini-code-assist]

Warning

You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again!

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀

[houseroad]

Could you list the test plan and test results?

[noiji]

Could you list the test plan and test results?

added it. thx @houseroad

[NickLucche]

lgtm

[houseroad]

Looks good. Thanks for the fix.

It will be good to add some unittest. (Feel free to do so in the following PR.)

[noiji]

It will be good to add some unittest. (Feel free to do so in the following PR.)

Will do in a follow-up PR. Thanks for the suggestion! @houseroad

[noiji]

Hi @houseroad @NickLucche all checks passed and PR is ready — could you help with the merge? Thanks!