fix

Author: ngxson

convert_hf_to_gguf.py

@@ -3067,6 +3067,11 @@ class Qwen3MoeModel(Qwen2MoeModel):
     model_arch = gguf.MODEL_ARCH.QWEN3MOE
 
 
+@ModelBase.register("Dots1ForCausalLM")
+class Dots1Model(Qwen2MoeModel):
+    model_arch = gguf.MODEL_ARCH.DOTS1
+
+
 @ModelBase.register("GPT2LMHeadModel")
 class GPT2Model(TextModel):
     model_arch = gguf.MODEL_ARCH.GPT2
@@ -5158,11 +5163,8 @@ def prepare_tensors(self):
                 raise ValueError(f"Unprocessed experts: {experts}")
 
 
-@ModelBase.register(
-    "DeepseekV2ForCausalLM",
-    "DeepseekV3ForCausalLM",
-    "Dots1ForCausalLM",
-)
+@ModelBase.register("DeepseekV2ForCausalLM")
+@ModelBase.register("DeepseekV3ForCausalLM")
 class DeepseekV2Model(TextModel):
     model_arch = gguf.MODEL_ARCH.DEEPSEEK2
 

gguf-py/gguf/constants.py

@@ -300,6 +300,7 @@ class MODEL_ARCH(IntEnum):
     QWEN2VL          = auto()
     QWEN3            = auto()
     QWEN3MOE         = auto()
+    DOTS1            = auto()
     PHI2             = auto()
     PHI3             = auto()
     PHIMOE           = auto()
@@ -580,6 +581,7 @@ class MODEL_TENSOR(IntEnum):
     MODEL_ARCH.QWEN2VL:          "qwen2vl",
     MODEL_ARCH.QWEN3:            "qwen3",
     MODEL_ARCH.QWEN3MOE:         "qwen3moe",
+    MODEL_ARCH.DOTS1:            "dots1",
     MODEL_ARCH.PHI2:             "phi2",
     MODEL_ARCH.PHI3:             "phi3",
     MODEL_ARCH.PHIMOE:           "phimoe",
@@ -1261,6 +1263,26 @@ class MODEL_TENSOR(IntEnum):
         MODEL_TENSOR.FFN_DOWN_EXP,
         MODEL_TENSOR.FFN_UP_EXP,
     ],
+    MODEL_ARCH.DOTS1: [
+        MODEL_TENSOR.TOKEN_EMBD,
+        MODEL_TENSOR.OUTPUT_NORM,
+        MODEL_TENSOR.OUTPUT,
+        MODEL_TENSOR.ATTN_NORM,
+        MODEL_TENSOR.ATTN_Q,
+        MODEL_TENSOR.ATTN_Q_NORM,
+        MODEL_TENSOR.ATTN_K,
+        MODEL_TENSOR.ATTN_K_NORM,
+        MODEL_TENSOR.ATTN_V,
+        MODEL_TENSOR.ATTN_OUT,
+        MODEL_TENSOR.FFN_NORM,
+        MODEL_TENSOR.FFN_GATE,
+        MODEL_TENSOR.FFN_UP,
+        MODEL_TENSOR.FFN_DOWN,
+        MODEL_TENSOR.FFN_GATE_INP,
+        MODEL_TENSOR.FFN_GATE_EXP,
+        MODEL_TENSOR.FFN_DOWN_EXP,
+        MODEL_TENSOR.FFN_UP_EXP,
+    ],
     MODEL_ARCH.PLAMO: [
         MODEL_TENSOR.TOKEN_EMBD,
         MODEL_TENSOR.OUTPUT_NORM,