PR #45441: fix(DSV3): parity between native DeepseekV3MoE and remote official implementation

• Repository: huggingface/transformers
• Author: casinca
• State: open | merged: False
• Link: https://github.com/huggingface/transformers/pull/45441

Description (problem / solution / changelog)

What does this PR do?

Please see fix #45440 for more details

Discussed with @vasqu

Also fixed via regen/inheritance:

• exaone-moe
• glm4-moe
• glm4v-moe
• glm4-moe-lite
• glm_moe_dsa
• nemotron_h
• solar_open

 

[!WARNING]

This test now fails (untrained, more likely to diverge) and needs to be regenerated.

https://github.com/huggingface/transformers/blob/5b565a589ca5935fbc7a3ea93b9b622bb41bd129/tests/models/deepseek_v3/test_modeling_deepseek_v3.py#L398-L409

 

So I reran the DSV3 forward pass to get the new hardcoded string for myself and retest but maybe this needs to be run and tested on your side too?

       EXPECTED_TEXT_COMPLETION = [
"Simply put, the theory of relativity states that aportersh455elike injection tactics-altitude蹲在那儿 >Loregefruitakosdeckingredientsuchtroni李世umontיםplicitlyShadowoldtriad Therapeutics不减-ste 的希望和价值 >kerretteylesheetzimnasium的品质 Talm",
"My favorite all time favorite condiment is ketchup. Lan overhead excite-ment好用>cileriaceaeagnainesogaslipadicSiggleESHalseawarriorsrattieri佐iented >Parrheta-counterousseanatysisoglCTSinkeheilbronnenlaceslide tactauralick",
       ]

 

• Also potential cascading effect with the inherited models too and their own def test_compile_static_cache(self): test?

 

I greedily flagged 3 other models with masked_fill(~score_mask.bool(), 0.0) but I'm not blindly touching these for now, need to verify their logic first (if using loss free load balancing) + what their remote are doing. If the remote is wrong, not sure we should change the native implementation or keep it wrong with the remote.

This is likely for a follow-up PR if these need to be fixed too.

Code Agent Policy

The Transformers repo is currently being overwhelmed by a large number of PRs and issue comments written by code agents. We are currently bottlenecked by our ability to review and respond to them. As a result, we ask that new users do not submit pure code agent PRs at this time. You may use code agents in drafting or to help you diagnose issues. We'd also ask autonomous "OpenClaw"-like agents not to open any PRs or issues for the moment.

PRs that appear to be fully agent-written will probably be closed without review, and we may block users who do this repeatedly or maliciously.

This is a rapidly-evolving situation that's causing significant shockwaves in the open-source community. As a result, this policy is likely to be updated regularly in the near future. For more information, please read CONTRIBUTING.md.

• I confirm that this is not a pure code agent PR.

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline, Pull Request section?
• Was this discussed/approved via a Github issue or the forum? Please add a link to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the documentation guidelines, and here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

<!-- Your PR will be replied to more quickly if you can figure out the right person to tag with @ If you know how to use git blame, that is the easiest way, otherwise, here is a rough guide of **who to tag**. Please tag fewer than 3 people. Models: - text models: @ArthurZucker @Cyrilvallez - vision models: @yonigozlan @molbap - audio models: @eustlb @ebezzam @vasqu - multimodal models: @zucchini-nlp - graph models: @clefourrier Library: - generate: @zucchini-nlp (visual-language models) or @gante (all others) - continuous batching: @remi-or @ArthurZucker @McPatate - pipelines: @Rocketknight1 - tokenizers: @ArthurZucker and @itazap - trainer: @SunMarc - attention: @vasqu @ArthurZucker @CyrilVallez - model loading (from pretrained, etc): @CyrilVallez - distributed: @3outeille @ArthurZucker - CIs: @ydshieh Integrations: - ray/raytune: @richardliaw, @amogkam - Big Model Inference: @SunMarc - quantization: @SunMarc - kernels: @drbh - peft: @BenjaminBossan @githubnemo Devices/Backends: - AMD ROCm: @ivarflakstad - Intel XPU: @IlyasMoutawwakil - Ascend NPU: @ivarflakstad Documentation: @stevhliu Research projects are not maintained and should be taken as is. -->

Changed files

• src/transformers/models/deepseek_v3/modeling_deepseek_v3.py (modified, +1/-1)
• src/transformers/models/deepseek_v3/modular_deepseek_v3.py (modified, +1/-1)
• src/transformers/models/exaone_moe/modeling_exaone_moe.py (modified, +1/-1)
• src/transformers/models/glm4_moe/modeling_glm4_moe.py (modified, +1/-1)
• src/transformers/models/glm4_moe_lite/modeling_glm4_moe_lite.py (modified, +1/-1)
• src/transformers/models/glm4v_moe/modeling_glm4v_moe.py (modified, +1/-1)
• src/transformers/models/glm_moe_dsa/modeling_glm_moe_dsa.py (modified, +1/-1)
• src/transformers/models/nemotron_h/modeling_nemotron_h.py (modified, +1/-1)
• src/transformers/models/solar_open/modeling_solar_open.py (modified, +1/-1)