Finding Your Base URL
The Base URL is the root address of the AI provider’s API. Batch Generate Text with any AI appends /chat/completions to it when making OpenAI-compatible requests.
If you select one of the nine built-in provider presets, the Base URL is filled in automatically. This page is mainly for custom endpoints or for understanding what the field means.
Base URLs for built-in providers
| Provider | Base URL |
|---|---|
| OpenAI | https://api.openai.com/v1 |
| DeepSeek | https://api.deepseek.com/v1 |
| OpenRouter | https://openrouter.ai/api/v1 |
| Mistral AI | https://api.mistral.ai/v1 |
| Together AI | https://api.together.xyz/v1 |
| xAI (Grok) | https://api.x.ai/v1 |
| Fireworks AI | https://api.fireworks.ai/inference/v1 |
| Cerebras | https://api.cerebras.ai/v1 |
| Google AI Studio | https://generativelanguage.googleapis.com/v1beta/openai/ |
Using a custom Base URL
Select Custom (OpenAI-Compatible) from the provider dropdown in AI Provider Settings, then enter your Base URL.
Common custom setups:
| Setup | Typical Base URL |
|---|---|
| Ollama (local) | http://localhost:11434/v1 |
| LM Studio (local) | http://localhost:1234/v1 |
| LocalAI | http://localhost:8080/v1 |
| Azure OpenAI | https://<your-resource>.openai.azure.com/openai/deployments/<deployment> |
| Any OpenAI-compatible server | Varies - check your provider’s docs |
The endpoint must support the OpenAI /chat/completions format. If your provider’s docs show a different path structure, use the part before /chat/completions as the Base URL.
Common mistakes
Trailing slash - most providers should not have a trailing slash. Use https://api.openai.com/v1, not https://api.openai.com/v1/. Google AI Studio’s preset is an exception because its OpenAI-compatible endpoint includes the final slash.
Including the endpoint path - the Base URL should usually stop before /chat/completions. The app adds that part automatically.
Wrong model for provider - a valid key can still fail if the model name does not exist for that provider.
Testing your connection
After entering a Base URL, model, and API key, click Save key, then Test connection in AI Provider Settings. The app sends a lightweight request to verify the key works before you start a batch.
If the test fails, see Troubleshooting for common error codes.