OpenRouter
OpenRouter is a unified API gateway that gives WednesdayAI access to hundreds of models from Anthropic, OpenAI, Google, Mistral, Meta, and others — all from one API key.When to use it
- Experiment with many models without managing separate keys
- Unified billing across providers
- Access to models not available in your region
Authentication
Create a key at openrouter.ai/keys, then onboard, or set it on the gateway host:Selecting a model
OpenRouter model references are prefixed withopenrouter/ and keep the upstream provider/model path:
| Model | Reference |
|---|---|
| Claude Sonnet 4.5 | openrouter/anthropic/claude-sonnet-4-5 |
| Claude Opus 4.6 | openrouter/anthropic/claude-opus-4-6 |
| Gemini 2.0 Flash | openrouter/google/gemini-2.0-flash-001 |
| Llama 3.3 70B | openrouter/meta-llama/llama-3.3-70b-instruct |
| DeepSeek R1 | openrouter/deepseek/deepseek-r1 |
When typing
/model in chat with an OpenRouter reference, include the full openrouter/ prefix — the reference is parsed by splitting on the first /, so the provider prefix is required for IDs that themselves contain /.openclaw models status to list models your key can access. WednesdayAI can also scan OpenRouter’s free-model catalog with openclaw models scan.
Troubleshooting
401 Unauthorized
401 Unauthorized
Model not available
Model not available
Not all models are available on all tiers. Check openrouter.ai/models for your key’s access.
Higher latency than a direct provider
Higher latency than a direct provider
OpenRouter adds a routing hop. Use it for experimentation; connect directly for latency-sensitive production.