[ERROR] engine: Failed to load model using from_pretrained #66
Description
Using CPU and docker. Build is successful, server comes up. However getting following error in the docker log, and, getting
"TTS engine model is not currently loaded or available" in server web interface:
2025-09-20 21:34:06 [INFO] engine: Attempting to load model directly using from_pretrained (expected from Hugging Face repository: ResembleAI/chatterbox or library default).
2025-09-20 21:34:11 [ERROR] engine: Failed to load model using from_pretrained (expected from 'ResembleAI/chatterbox' or library default): Error while deserializing header: MetadataIncompleteBuffer
Traceback (most recent call last):
File "/app/engine.py", line 172, in load_model
chatterbox_model = ChatterboxTTS.from_pretrained(device=model_device)
File "/usr/local/lib/python3.10/dist-packages/chatterbox/tts.py", line 180, in from_pretrained
return cls.from_local(Path(local_path).parent, device)
File "/usr/local/lib/python3.10/dist-packages/chatterbox/tts.py", line 145, in from_local
t3_state = load_file(ckpt_dir / "t3_cfg.safetensors")
File "/usr/local/lib/python3.10/dist-packages/safetensors/torch.py", line 313, in load_file
with safe_open(filename, framework="pt", device=device) as f:
safetensors_rust.SafetensorError: Error while deserializing header: MetadataIncompleteBuffer
2025-09-20 21:34:11 [CRITICAL] server: CRITICAL: TTS Model failed to load on startup. Server might not function correctly.