diff --git a/extensions_built_in/diffusion_models/qwen_image/qwen_image.py b/extensions_built_in/diffusion_models/qwen_image/qwen_image.py
index b4491962..0163dc19 100644
--- a/extensions_built_in/diffusion_models/qwen_image/qwen_image.py
+++ b/extensions_built_in/diffusion_models/qwen_image/qwen_image.py
@@ -256,8 +256,11 @@ class QwenImageModel(BaseModel):
         latent_model_input = latent_model_input.permute(0, 2, 4, 1, 3, 5)
         latent_model_input = latent_model_input.reshape(batch_size, (height // 2) * (width // 2), num_channels_latents * 4)
         
-        prompt_embeds_mask = text_embeddings.attention_mask.to(self.device_torch, dtype=torch.int64)
-        
+        # make txt_seq_lens match the actual encoder_hidden_states length, and clamp the mask ---
+        seq_len = text_embeddings.text_embeds.shape[1]
+        prompt_embeds_mask = text_embeddings.attention_mask.to(self.device_torch, dtype=torch.int64)[:, :seq_len]
+        txt_seq_lens = [seq_len] * batch_size
+
         img_shapes = [(1, height // 2, width  // 2)] * batch_size
 
         noise_pred = self.transformer(
@@ -267,12 +270,11 @@ class QwenImageModel(BaseModel):
             encoder_hidden_states=text_embeddings.text_embeds.to(self.device_torch, self.torch_dtype),
             encoder_hidden_states_mask=prompt_embeds_mask,
             img_shapes=img_shapes,
-            txt_seq_lens=prompt_embeds_mask.sum(dim=1).tolist(),
+            txt_seq_lens=txt_seq_lens,
             return_dict=False,
             **kwargs,
         )[0]
         
-        # unpack the noise prediction
         noise_pred = noise_pred.view(batch_size, height // 2, width // 2, num_channels_latents, 2, 2)
         noise_pred = noise_pred.permute(0, 3, 1, 4, 2, 5)
         noise_pred = noise_pred.reshape(batch_size, num_channels_latents, height, width)