Skip to main content

​
Venice AI

Venice AI is a privacy-focused inference provider. Prompts and responses are not logged or used for training, making it suitable for sensitive workloads.

​
Prerequisites

Create an API key at venice.ai and set it on the gateway host: 
echo 'VENICE_API_KEY=your-key-here' >> ~/.openclaw/.env

​
Configuration

{
  models: {
    providers: {
      venice: {
        apiKey: "VENICE_API_KEY", // references the env var
        models: [
          "llama-3.3-70b",
          "mistral-31-24b",
        ],
      },
    },
  },
}
apiKey is the name of the environment variable, not the key value itself. Keep secrets in .env, not in committed config files.

​
Selecting a model

{
  agents: {
    defaults: {
      model: { primary: "venice/llama-3.3-70b" },
    },
  },
}
ModelNotes
llama-3.3-70bLlama 3.3 70B — strong general-purpose
mistral-31-24bMistral 3.1 24B — efficient mid-size model
Venice does not log prompts or responses. For compliance-sensitive deployments, verify Venice’s current data handling policy at venice.ai/privacy.
Run openclaw models list --provider venice for the full current catalog.

​
Troubleshooting

The API key is invalid or revoked. Generate a new one at venice.ai and update ~/.openclaw/.env.
Venice’s model catalog may differ from other providers. Check the Venice dashboard for currently available models.