Oh, forgot to commit

convert_hf_to_gguf.py

@@ -3760,6 +3760,8 @@ class Qwen3NextModel(Qwen3MoeModel):
         self.gguf_writer.add_ssm_group_count(self.find_hparam(["linear_num_key_heads"]))
         self.gguf_writer.add_ssm_time_step_rank(self.find_hparam(["linear_num_value_heads"]))
         self.gguf_writer.add_ssm_inner_size(self.find_hparam(['linear_value_head_dim']) * self.find_hparam(['linear_num_value_heads']))
+        rope_dim = self.hparams["hidden_size"] // self.hparams["num_attention_heads"]
+        self.gguf_writer.add_rope_dimension_count(int(rope_dim * self.hparams.get("partial_rotary_factor", 0.25)))
 
     def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iterable[tuple[str, Tensor]]:
         if name.startswith("mtp"):