Venice AI Provider Setup for OpenClaw

Venice AI offers inference with privacy as a priority: open models run without any logging, and Claude, GPT, Gemini, and Grok are reachable through anonymized proxies. Every endpoint follows the OpenAI-compatible format (/v1).

Privacy modes

Warning

Anonymized models fall short of full privacy. Venice strips metadata prior to forwarding, yet the upstream provider (OpenAI, Anthropic, Google, xAI) still handles the request. When absolute privacy matters, go with Private models.

Getting started

Install the plugin

openclaw plugins install @openclaw/venice-provider

Get your API key

1. Create an account at venice.ai
2. Navigate to Settings > API Keys > Create new key
3. Grab your API key (format: vapi_xxxxxxxxxxxx)

Configure OpenClaw

Interactive (recommended)

openclaw onboard --auth-choice venice-api-key

Asks for the API key (or reuses an existing VENICE_API_KEY), shows the available Venice models, and designates your default model.

Environment variable

export VENICE_API_KEY="vapi_xxxxxxxxxxxx"

Non-interactive

openclaw onboard --non-interactive --accept-risk --skip-health \
  --auth-choice venice-api-key \
  --venice-api-key "vapi_xxxxxxxxxxxx"

Verify setup

openclaw agent --model venice/zai-org-glm-4.7 --message "Hello, are you working?"

Model selection

• Default: venice/zai-org-glm-4.7 (private reasoning).
• Top anonymized pick: venice/claude-opus-5.

openclaw models set venice/zai-org-glm-4.7
openclaw models list --all --provider venice

Launching openclaw configure and choosing Model/auth provider > Venice AI is another route.

Tip

Use case	Model	Why
General chat (default)	zai-org-glm-4.7	Venice live default trait
Best overall quality	claude-opus-5	Current promoted anonymized Opus model
Privacy + coding	qwen3-coder-480b-a35b-instruct-turbo	Private coding model with large context
Fast + cheap	google-gemma-4-31b-it	Low-cost promoted private vision model
Complex private tasks	deepseek-v3.2	Promoted private reasoning model
Uncensored	venice-uncensored-1-2	Current uncensored Venice model

Built-in catalog (16 visible models)

Private models (10) — fully private, no logging

Anonymized models (6) — via Venice proxy

Deprecated compatibility rows (3) — hidden from pickers

Venice models powered by Grok (grok-4-3 and others) receive the identical tool-schema compatibility patch as the native xAI provider, because their upstream tool-call format is shared.

Model discovery

The catalog listed above starts as a manifest-backed seed set. During runtime, OpenClaw pulls fresh data from the Venice /models API, reverting to the seed set when the API cannot be reached. The /models endpoint is public, so listing needs no authentication, yet running inference still demands a valid API key.

Retired model IDs might keep working as provider-owned aliases on Venice. OpenClaw's catalog only presents the canonical model IDs that /models returns.

DeepSeek V4 replay behavior

If Venice serves DeepSeek V4 models like deepseek-v4-pro or deepseek-v4-flash, OpenClaw automatically supplies the reasoning_content replay field on assistant messages whenever Venice leaves it out, and it removes thinking/reasoning/reasoning_effort from the request body, since Venice rejects DeepSeek's native thinking control for these models. This replay adjustment is distinct from the thinking controls found in the native DeepSeek provider.

Streaming and tool support

Pricing

Venice operates on a credit-based model. Anonymous models are priced close to direct API rates, with a modest Venice surcharge added. For up-to-date pricing, refer to venice.ai/pricing.

OpenClaw relies on the Venice plugin manifest for pricing of bundled models, even when those models show up through live discovery. Models found via discovery that aren't in the manifest retain zero estimates until pricing gets set; a zero value doesn't imply the model is free.

Explicitly configured model prices take priority, including zero values. Onboarding keeps existing model entries intact rather than overwriting their prices. If an older setup has outdated zero rates, back up your configuration and adjust only the relevant models.providers.venice.models[].cost fields to reflect current pricing. Keep any deliberate custom prices as they are. See Token use and costs for details on pricing units and usage reporting.

Usage examples

# Default private model
openclaw agent --model venice/zai-org-glm-4.7 --message "Quick health check"

# Claude Opus via Venice (anonymized)
openclaw agent --model venice/claude-opus-5 --message "Summarize this task"

# Uncensored model
openclaw agent --model venice/venice-uncensored-1-2 --message "Draft options"

# Vision model with image
openclaw agent --model venice/qwen3-vl-235b-a22b --message "Review attached image"

# Coding model
openclaw agent --model venice/qwen3-coder-480b-a35b-instruct-turbo --message "Refactor this function"

Troubleshooting

API key not recognized

openclaw models list --provider venice

Make sure the API key is set up and begins with vapi_; never print or share its value.

Model not available

Execute openclaw models list --all --provider venice to list the models currently available; this catalog shifts as Venice introduces or removes models.

Connection issues

The Venice API endpoint is https://api.venice.ai/api/v1. Verify your network permits HTTPS traffic to that host.

Note

Additional assistance: Troubleshooting and FAQ.

Advanced configuration

Config file example

{
  env: { vars: { VENICE_API_KEY: "vapi_..." } },
  agents: { defaults: { model: { primary: "venice/zai-org-glm-4.7" } } },
  models: {
    mode: "merge",
    providers: {
      venice: {
        baseUrl: "https://api.venice.ai/api/v1",
        apiKey: "${VENICE_API_KEY}",
        api: "openai-completions",
        models: [
          {
            id: "zai-org-glm-4.7",
            name: "GLM 4.7",
            reasoning: true,
            input: ["text"],
            cost: { input: 0.55, output: 2.65, cacheRead: 0.11, cacheWrite: 0 },
            contextWindow: 198000,
            maxTokens: 16384,
          },
        ],
      },
    },
  },
}

Related

• Model selection, How to pick providers, model references, and failover settings.

• Venice AI, Venice AI's homepage and account registration.

• API documentation, Venice API reference and developer guides.

• Pricing, Current Venice credit rates and subscription plans.