docker-compose cpu only fails.

[haydonryan]

Thanks for putting this out there.
I am trying to run this on a machine that does not have a gpu, nor any nvidia parts. I'm using the docker-compose-cpu file.

Few issues here.

1. docker has deprecated the version tags - probably a good idea to remove them.
   

   Chatterbox-TTS-Server/docker-compose-cpu.yml

   Lines 1 to 2 in d527f9f

   	version: '3.8'

2. config.yml contains device: cuda which is incompatable with cpu.
   

   Chatterbox-TTS-Server/config.yaml

   Line 14 in d527f9f

   device: cuda

3. it's flapping on my machine.

2026-02-14 20:49:21 [INFO] __main__: Starting TTS Server directly on http://0.0.0.0:8004
2026-02-14 20:49:21 [INFO] __main__: API documentation will be available at http://0.0.0.0:8004/docs
2026-02-14 20:49:21 [INFO] __main__: Web UI will be available at http://0.0.0.0:8004/
2026-02-14 20:49:21 [INFO] server: TTS Server: Initializing application...
2026-02-14 20:49:21 [INFO] server: Configuration loaded. Log file at: /app/logs/tts_server.log
2026-02-14 20:49:21 [INFO] engine: CPU device explicitly requested in config. Using CPU.
2026-02-14 20:49:21 [INFO] engine: Final device selection: cpu
2026-02-14 20:49:21 [INFO] engine: Model selector from config: 'chatterbox-turbo'
2026-02-14 20:49:21 [INFO] engine: Model selector 'chatterbox-turbo' resolved to Turbo model (ChatterboxTurboTTS)
2026-02-14 20:49:21 [INFO] engine: Initializing ChatterboxTurboTTS on device 'cpu'...
2026-02-14 20:49:21 [INFO] engine: Model type: turbo
2026-02-14 20:49:21 [INFO] engine: Turbo model supports paralinguistic tags: ['laugh', 'chuckle', 'sigh', 'gasp', 'cough', 'clear throat', 'sniff', 'groan', 'shush']

$ docker-compose up
Attaching to chatterbox-tts-server-1
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | ==========
chatterbox-tts-server-1  | == CUDA ==
chatterbox-tts-server-1  | ==========
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | CUDA Version 12.8.1
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | Container image Copyright (c) 2016-2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | This container image and its contents are governed by the NVIDIA Deep Learning Container License.
chatterbox-tts-server-1  | By pulling and using the container, you accept the terms and conditions of this license:
chatterbox-tts-server-1  | https://developer.nvidia.com/ngc/nvidia-deep-learning-container-license
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | A copy of this license is made available in this container at /NGC-DL-CONTAINER-LICENSE for your convenience.
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | WARNING: The NVIDIA Driver was not detected.  GPU functionality will not be available.
chatterbox-tts-server-1  |    Use the NVIDIA Container Toolkit to start this container with GPU support; see
chatterbox-tts-server-1  |    https://docs.nvidia.com/datacenter/cloud-native/ .
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] __main__: Starting TTS Server directly on http://0.0.0.0:8004
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] __main__: API documentation will be available at http://0.0.0.0:8004/docs
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] __main__: Web UI will be available at http://0.0.0.0:8004/
chatterbox-tts-server-1  | INFO:     Started server process [1]
chatterbox-tts-server-1  | INFO:     Waiting for application startup.
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] server: TTS Server: Initializing application...
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] server: Configuration loaded. Log file at: /app/logs/tts_server.log
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] engine: CPU device explicitly requested in config. Using CPU.
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] engine: Final device selection: cpu
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] engine: Model selector from config: 'chatterbox-turbo'
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] engine: Model selector 'chatterbox-turbo' resolved to Turbo model (ChatterboxTurboTTS)
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] engine: Initializing ChatterboxTurboTTS on device 'cpu'...
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] engine: Model type: turbo
chatterbox-tts-server-1  | 2026-02-14 21:03:54 [INFO] engine: Turbo model supports paralinguistic tags: ['laugh', 'chuckle', 'sigh', 'gasp', 'cough', 'clear throat', 'sniff', 'groan', 'shush']
chatterbox-tts-server-1 exited with code 137 (restarting)
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | ==========
chatterbox-tts-server-1  | == CUDA ==
chatterbox-tts-server-1  | ==========
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | CUDA Version 12.8.1
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | Container image Copyright (c) 2016-2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | This container image and its contents are governed by the NVIDIA Deep Learning Container License.
chatterbox-tts-server-1  | By pulling and using the container, you accept the terms and conditions of this license:
chatterbox-tts-server-1  | https://developer.nvidia.com/ngc/nvidia-deep-learning-container-license
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | A copy of this license is made available in this container at /NGC-DL-CONTAINER-LICENSE for your convenience.
chatterbox-tts-server-1  |
chatterbox-tts-server-1  | WARNING: The NVIDIA Driver was not detected.  GPU functionality will not be available.
chatterbox-tts-server-1  |    Use the NVIDIA Container Toolkit to start this container with GPU support; see
chatterbox-tts-server-1  |    https://docs.nvidia.com/datacenter/cloud-native/ .
chatterbox-tts-server-1  |

It starts, hits high cpu, then restarts. and repeats. I also note that it comes up saying Cuda as it's still using the Cuda Docker images. (note i copied the docker-compose-cpu.yml).

Not sure why it's exiting - it's there's no errors coming up.

[devnen]

This is now fixed in commit a67348a. Three changes address the problems you reported:

1. Lightweight CPU image: New Dockerfile.cpu based on python:3.10-slim instead of the 4GB+ nvidia/cuda base image. docker-compose-cpu.yml now uses this. No more CUDA warnings on CPU-only builds.
2. Deprecated version tags: Removed version: '3.8' from all docker-compose files.
3. config.yaml default: Changed device: cuda to device: auto — the engine auto-detects CUDA, MPS, or CPU.

To update:

git pull
docker compose -f docker-compose-cpu.yml build --no-cache
docker compose -f docker-compose-cpu.yml up

Note: Exit code 137 (OOM) during model loading is typically caused by insufficient RAM for the model on CPU. Chatterbox Turbo requires ~4-6GB RAM for inference.