Upgrade to vllm 0.17.0 corex v4.1 overlay

vllm/model_executor/models/registry.py

@@ -75,12 +75,14 @@ _TEXT_GENERATION_MODELS = {
     "AquilaForCausalLM": ("llama", "LlamaForCausalLM"),  # AquilaChat2
     "ArceeForCausalLM": ("arcee", "ArceeForCausalLM"),
     "ArcticForCausalLM": ("arctic", "ArcticForCausalLM"),
+    "AXK1ForCausalLM": ("AXK1", "AXK1ForCausalLM"),
     # baichuan-7b, upper case 'C' in the class name
     "BaiChuanForCausalLM": ("baichuan", "BaiChuanForCausalLM"),
     # baichuan-13b, lower case 'c' in the class name
     "BaichuanForCausalLM": ("baichuan", "BaichuanForCausalLM"),
     "BailingMoeForCausalLM": ("bailing_moe", "BailingMoeForCausalLM"),
     "BailingMoeV2ForCausalLM": ("bailing_moe", "BailingMoeV2ForCausalLM"),
+    "BailingMoeV2_5ForCausalLM": ("bailing_moe_linear", "BailingMoeV25ForCausalLM"),
     "BambaForCausalLM": ("bamba", "BambaForCausalLM"),
     "BloomForCausalLM": ("bloom", "BloomForCausalLM"),
     "ChatGLMModel": ("chatglm", "ChatGLMForCausalLM"),
@@ -259,6 +261,10 @@ _EMBEDDING_MODELS = {
     "OpsColQwen3Model": ("colqwen3", "ColQwen3Model"),
     "Qwen3VLNemotronEmbedModel": ("colqwen3", "ColQwen3Model"),
     "SiglipModel": ("siglip", "SiglipEmbeddingModel"),
+    "LlamaNemotronVLModel": (
+        "nemotron_vl",
+        "LlamaNemotronVLForEmbedding",
+    ),
     # Technically Terratorch models work on images, both in
     # input and output. I am adding it here because it piggy-backs on embedding
     # models for the time being.
@@ -278,6 +284,10 @@ _CROSS_ENCODER_MODELS = {
         "llama",
         "LlamaBidirectionalForSequenceClassification",
     ),
+    "LlamaNemotronVLForSequenceClassification": (
+        "nemotron_vl",
+        "LlamaNemotronVLForSequenceClassification",
+    ),
     "ModernBertForSequenceClassification": (
         "modernbert",
         "ModernBertForSequenceClassification",
@@ -331,6 +341,10 @@ _MULTIMODAL_MODELS = {
         "ernie45_vl",
         "Ernie4_5_VLMoeForConditionalGeneration",
     ),
+    "FireRedASR2ForConditionalGeneration": (
+        "fireredasr2",
+        "FireRedASR2ForConditionalGeneration",
+    ),
     "FunASRForConditionalGeneration": ("funasr", "FunASRForConditionalGeneration"),  # noqa: E501
     "FunAudioChatForConditionalGeneration": (
         "funaudiochat",
@@ -506,6 +520,7 @@ _MULTIMODAL_MODELS = {
 }
 
 _SPECULATIVE_DECODING_MODELS = {
+    "ExtractHiddenStatesModel": ("extract_hidden_states", "ExtractHiddenStatesModel"),
     "MiMoMTPModel": ("mimo_mtp", "MiMoMTP"),
     "EagleLlamaForCausalLM": ("llama_eagle", "EagleLlamaForCausalLM"),
     "EagleLlama4ForCausalLM": ("llama4_eagle", "EagleLlama4ForCausalLM"),