Merge branch 'samuraieng/sarashina22vl-01' of github.com:samuraieng/llama.cpp into samuraieng/sarashina22vl-01

Author: samuraieng

convert_hf_to_gguf.py

@@ -13551,6 +13551,26 @@ def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iter
         yield from super().modify_tensors(data_torch, name, bid)
 
 
+@ModelBase.register("Sarashina2VisionForCausalLM")
+class Sarashina2VLTextModel(LlamaModel):
+    model_arch = gguf.MODEL_ARCH.LLAMA
+
+    def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None):
+        if name.startswith("llm."):
+            name = name.replace("llm.", "", 1)
+        elif name.startswith("norm.") or name.startswith("visual."):
+            return
+
+        yield from super().modify_tensors(data_torch, name, bid)
+
+
+@ModelBase.register("Sarashina2VisionForCausalLM")
+class Sarashina2VLVisionModel(Qwen2VLVisionModel):
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.global_config['model_type'] = "qwen2_vl"
+
+
 ###### CONVERSION LOGIC ######
 
 
@@ -13807,7 +13827,7 @@ def get_model_architecture(hparams: dict[str, Any], model_type: ModelType) -> st
     # Step3-VL keeps text config under text_config but uses a custom top-level architecture.
     # For text conversion we route to a dedicated text-only class.
     # TODO: refactor this later to avoid adding exception here
-    if model_type == ModelType.TEXT and arch == "StepVLForConditionalGeneration":
+    if model_type == ModelType.TEXT and arch in ("StepVLForConditionalGeneration", "Sarashina2VisionForCausalLM"):
         return arch
 
     # if "architectures" is found in the sub-config, use that instead