[model] Add support for openPangu_Ultra_MoE (#27521)

Signed-off-by: yuantao <2422264527@qq.com>
Signed-off-by: yt0428 <51468697+yt0428@users.noreply.github.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
This commit is contained in:
yt0428
2025-11-04 08:17:20 -08:00
committed by GitHub
co-authored by Jee Jee Li
parent 5fd8f02ea9
commit 05cae69f0f
8 changed files with 1379 additions and 1 deletions
+3
View File
@@ -149,6 +149,8 @@ _TEXT_GENERATION_MODELS = {
"OPTForCausalLM": ("opt", "OPTForCausalLM"),
"OrionForCausalLM": ("orion", "OrionForCausalLM"),
"OuroForCausalLM": ("ouro", "OuroForCausalLM"),
"PanguEmbeddedForCausalLM": ("openpangu", "PanguEmbeddedForCausalLM"),
"PanguUltraMoEForCausalLM": ("openpangu", "PanguUltraMoEForCausalLM"),
"PersimmonForCausalLM": ("persimmon", "PersimmonForCausalLM"),
"PhiForCausalLM": ("phi", "PhiForCausalLM"),
"Phi3ForCausalLM": ("phi3", "Phi3ForCausalLM"),
@@ -406,6 +408,7 @@ _SPECULATIVE_DECODING_MODELS = {
"LongCatFlashMTPModel": ("longcat_flash_mtp", "LongCatFlashMTP"),
"Glm4MoeMTPModel": ("glm4_moe_mtp", "Glm4MoeMTP"),
"MedusaModel": ("medusa", "Medusa"),
"OpenPanguMTPModel": ("openpangu_mtp", "OpenPanguMTP"),
"Qwen3NextMTP": ("qwen3_next_mtp", "Qwen3NextMTP"),
# Temporarily disabled.
# # TODO(woosuk): Re-enable this once the MLP Speculator is supported in V1.