Skip to content

dstackai/openclaw-demo

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Deploying OpenClaw (ClawdBot) on GPU clouds with dstack

Prerequisites

  1. Install dstack
  2. Set up a dstack gateway (already set up for you if you sign up with dstack Sky)

Deploy a model (Qwen 32B)

type: service name: qwen32 image: lmsysorg/sglang:latest env: - HF_TOKEN - MODEL_ID=deepseek-ai/DeepSeek-R1-Distill-Qwen-32B commands: # Launch SGLang server - |  python3 -m sglang.launch_server \  --model-path $MODEL_ID \  --host 0.0.0.0 \  --port 8000  port: 8000 model: deepseek-ai/DeepSeek-R1-Distill-Qwen-32B # Uncomment to use allow spot instances #spot_policy: auto resources: gpu: H100 disk: 200GB
export HF_TOKEN=... dstack apply -f qwen32.dstack.yml

Once the model is deployed, the OpenAI-compatible base URL will be avaiable at https://qwen32.<your gateway domain>/v1.

Deploy OpenClaw

type: service name: openclaw env: - MODEL_BASE_URL - DSTACK_TOKEN commands: # System dependencies - sudo apt update - sudo apt install -y ca-certificates curl gnupg # Node.js & openclaw installation - curl -fsSL https://deb.nodesource.com/setup_24.x | sudo -E bash - - sudo apt install -y nodejs - curl -fsSL https://openclaw.bot/install.sh | bash -s -- --no-onboard # Model configuration - |  openclaw config set models.providers.custom '{  "baseUrl": "'"$MODEL_BASE_URL"'",  "apiKey": "'"$DSTACK_TOKEN"'",  "api": "openai-completions",  "models": [  {  "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",  "name": "DeepSeek-R1-Distill-Qwen-32B",  "reasoning": true,  "input": ["text"],  "cost": {"input": 0, "output": 0, "cacheRead": 0, "cacheWrite": 0},  "contextWindow": 128000,  "maxTokens": 512  }  ]  }' --json   # Set active model & gateway settings - openclaw models set custom/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B - openclaw config set gateway.mode local - openclaw config set gateway.auth.mode token - openclaw config set gateway.auth.token "$DSTACK_TOKEN" - openclaw config set gateway.controlUi.allowInsecureAuth true - openclaw config set gateway.trustedProxies '["127.0.0.1"]' # Start service - openclaw gateway port: 18789 auth: false resources: cpu: 2
export MODEL_BASE_URL=<your model endpoint> DSTACK_TOKEN=<your dstack token> dstack apply -f openclaw.dstack.yml

The Moltbot UI will be available at https://openclaw.<your gateway domain>/chat.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages