Update README.md

README.md

@@ -947,9 +947,9 @@ Click here to try the online demo of [MiniCPM-o 2.6](https://minicpm-omni-webdem
 Inference using Huggingface transformers on NVIDIA GPUs. Please ensure that `transformers==4.44.2` is installed, as other versions may have compatibility issues. We are investigating this issue. Requirements tested on python 3.10：
 ```
 Pillow==10.1.0
-torch==2.2.0
-torchaudio==2.2.0
-torchvision==0.17.0
+torch==2.3.1
+torchaudio==2.3.1
+torchvision==0.18.1
 transformers==4.44.2
 librosa==0.9.0
 soundfile==0.12.1
@@ -986,8 +986,13 @@ tokenizer = AutoTokenizer.from_pretrained('openbmb/MiniCPM-o-2_6', trust_remote_
 
 # In addition to vision-only mode, tts processor and vocos also needs to be initialized
 model.init_tts()
+```
+
+If you are using an older version of PyTorch, you might encounter this issue `"weight_norm_fwd_first_dim_kernel" not implemented for 'BFloat16'`, Please convert the TTS to float32 type.
+```python
 model.tts.float()
 ```
+
 ### Omni mode
 we provide two inference modes: chat and streaming