Skip to content

[model] Add support for openPangu_Ultra_MoE #27521

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 9 commits into from
Nov 4, 2025
+1,379 −1
Merged

[model] Add support for openPangu_Ultra_MoE #27521

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
2aa3938
[model] Add support for openpangu
yt0428 Oct 26, 2025
8cc5682
Doc fix, delete extra data column in supported_models.md for PanguUlt…
yt0428 Oct 26, 2025
433a7ee
Fix reviews
yt0428 Oct 30, 2025
8ae429b
Add links for model
yt0428 Oct 30, 2025
307b825
Refactor openpangu.py to further add support for openPangu-Embedded m…
yt0428 Oct 30, 2025
6191527
Merge branch 'main' into support_openpangu
jeejeelee Oct 30, 2025
9804db4
Merge branch 'main' into support_openpangu
yt0428 Oct 31, 2025
5a959b6
Update huggingface model links
yt0428 Nov 3, 2025
5a690e4
Merge branch 'main' into support_openpangu
jeejeelee Nov 4, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions docs/models/supported_models.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -404,6 +404,8 @@ th {
| `OPTForCausalLM` | OPT, OPT-IML | `facebook/opt-66b`, `facebook/opt-iml-max-30b`, etc. | ✅︎ | ✅︎ |
| `OrionForCausalLM` | Orion | `OrionStarAI/Orion-14B-Base`, `OrionStarAI/Orion-14B-Chat`, etc. | | ✅︎ |
| `OuroForCausalLM` | ouro | `ByteDance/Ouro-1.4B`, `ByteDance/Ouro-2.6B`, etc. | ✅︎ | |
| `PanguEmbeddedForCausalLM` |openPangu-Embedded-7B | `FreedomIntelligence/openPangu-Embedded-7B-V1.1` | ✅︎ | ✅︎ |
| `PanguUltraMoEForCausalLM` |openpangu-ultra-moe-718b-model | `FreedomIntelligence/openPangu-Ultra-MoE-718B-V1.1` | ✅︎ | ✅︎ |
| `PhiForCausalLM` | Phi | `microsoft/phi-1_5`, `microsoft/phi-2`, etc. | ✅︎ | ✅︎ |
| `Phi3ForCausalLM` | Phi-4, Phi-3 | `microsoft/Phi-4-mini-instruct`, `microsoft/Phi-4`, `microsoft/Phi-3-mini-4k-instruct`, `microsoft/Phi-3-mini-128k-instruct`, `microsoft/Phi-3-medium-128k-instruct`, etc. | ✅︎ | ✅︎ |
| `PhiMoEForCausalLM` | Phi-3.5-MoE | `microsoft/Phi-3.5-MoE-instruct`, etc. | ✅︎ | ✅︎ |
Expand Down
13 changes: 13 additions & 0 deletions tests/models/registry.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -363,13 +363,26 @@ def check_available_online(
"Olmo2ForCausalLM": _HfExamplesInfo("allenai/OLMo-2-0425-1B"),
"Olmo3ForCausalLM": _HfExamplesInfo("shanearora/2025-sep-a-base-model"),
"OlmoeForCausalLM": _HfExamplesInfo("allenai/OLMoE-1B-7B-0924-Instruct"),
"OpenPanguMTPModel": _HfExamplesInfo(
"FreedomIntelligence/openPangu-Ultra-MoE-718B-V1.1",
trust_remote_code=True,
is_available_online=False,
),
"OPTForCausalLM": _HfExamplesInfo(
"facebook/opt-125m", {"1b": "facebook/opt-iml-max-1.3b"}
),
"OrionForCausalLM": _HfExamplesInfo(
"OrionStarAI/Orion-14B-Chat", trust_remote_code=True
),
"OuroForCausalLM": _HfExamplesInfo("ByteDance/Ouro-1.4B", trust_remote_code=True),
"PanguEmbeddedForCausalLM": _HfExamplesInfo(
"FreedomIntelligence/openPangu-Embedded-7B-V1.1", trust_remote_code=True
),
"PanguUltraMoEForCausalLM": _HfExamplesInfo(
"FreedomIntelligence/openPangu-Ultra-MoE-718B-V1.1",
trust_remote_code=True,
is_available_online=False,
),
"PersimmonForCausalLM": _HfExamplesInfo("adept/persimmon-8b-chat"),
"PhiForCausalLM": _HfExamplesInfo("microsoft/phi-2"),
"Phi3ForCausalLM": _HfExamplesInfo("microsoft/Phi-3-mini-4k-instruct"),
Expand Down
3 changes: 3 additions & 0 deletions vllm/config/model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1231,6 +1231,8 @@ def is_deepseek_mla(self) -> bool:
"kimi_k2",
"kimi_linear",
"longcat_flash",
"pangu_ultra_moe",
"pangu_ultra_moe_mtp",
):
return self.hf_text_config.kv_lora_rank is not None
elif self.hf_text_config.model_type == "eagle":
Expand Down Expand Up @@ -1379,6 +1381,7 @@ def get_layers_start_end_indices(
or self.hf_config.model_type == "glm4_moe_mtp"
or self.hf_config.model_type == "ernie_mtp"
or self.hf_config.model_type == "qwen3_next_mtp"
or self.hf_config.model_type == "pangu_ultra_moe_mtp"
):
total_num_hidden_layers = getattr(
self.hf_text_config, "num_nextn_predict_layers", 0
Expand Down
9 changes: 9 additions & 0 deletions vllm/config/speculative.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,6 +41,7 @@
"qwen3_next_mtp",
"mimo_mtp",
"longcat_flash_mtp",
"pangu_ultra_moe_mtp",
"mtp",
"suffix",
]
Expand All @@ -51,6 +52,7 @@
"ernie_mtp",
"qwen3_next_mtp",
"longcat_flash_mtp",
"pangu_ultra_moe_mtp",
)


Expand Down Expand Up @@ -179,6 +181,13 @@ def hf_config_override(hf_config: PretrainedConfig) -> PretrainedConfig:
hf_config.update(
{"n_predict": n_predict, "architectures": ["DeepSeekMTPModel"]}
)
if hf_config.model_type in ("pangu_ultra_moe"):
hf_config.model_type = "pangu_ultra_moe_mtp"
if hf_config.model_type == "pangu_ultra_moe_mtp":
n_predict = getattr(hf_config, "num_nextn_predict_layers", None)
hf_config.update(
{"n_predict": n_predict, "architectures": ["OpenPanguMTPModel"]}
)
Comment on lines +184 to +190
Copy link
Member

@hmellor hmellor Oct 29, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could this override be done in vllm/model_executor/models/openpangu.py and vllm/model_executor/models/openpangu_mtp.py instead of putting model specific config overrides in the global configs?

Copy link
Member

@hmellor hmellor Oct 29, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm I see this is currently done for quite a few models... We should do this in a follow up

Copy link
Contributor Author

@yt0428 yt0428 Oct 30, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I follow the common practice (like qwen3_next_mtp, longcat_flash_mtp) and place the override for mtp in vllm/config/speculative.py. I also thought about move the override to the modeling definition file, but it seems the initialization work flow for mtp makes it infeasible. Could you please provide some suggestions on the implementation?

Copy link
Member

@hmellor hmellor Oct 30, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

For this PR, please follow the existing pattern as you have already done. Refactoring MTP config is a separate task.

Copy link
Contributor Author

@yt0428 yt0428 Oct 30, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Cool! Looking forward to it.


if hf_config.architectures[0] == "MiMoForCausalLM":
hf_config.model_type = "mimo_mtp"
Expand Down
Loading