OpenRouter api Integration Guide

Why Switch Your OpenRouter API Endpoint?

Many developers encounter rate limits or content filters when using free tiers of popular uncensored models. You might see errors like "no endpoints found" or sudden throttling during peak hours. Our service isolates the Dolphin model on dedicated servers to ensure high availability. This eliminates the variability of shared routing infrastructure. You get direct access to a single, powerful model without training data interference. The API remains fully OpenAI-compatible, meaning your existing client code works with zero structural changes. This approach guarantees predictable latency for roleplay or unrestricted text generation. You avoid the frustration of model routing issues that plague multi-model aggregators.

Step 1: Get Your Uncensored API Key

You can generate your API key immediately by signing up with your email or Google account. No phone number verification is required for this process. Once registered, your key is displayed instantly on the dashboard. Each account is limited to one active key to ensure secure tracking. If you need a new key, the old one is replaced immediately. This streamlined signup process allows you to start testing your integration within minutes. The key grants access to the uncensored model without any subscription fees. You only pay for the tokens you actually consume.

Step 2: Update Your Base URL

To migrate your application, you must update the base URL in your OpenAI-compatible client configuration. Point your SDK to our dedicated endpoint at https://api.openrouterdolphinai.com/v1. This URL works with any standard OpenAI SDK implementation. You do not need to change your model ID; use uncensored as the model identifier. This simple change directs your requests to our isolated servers. It ensures that your traffic is not mixed with other users of shared aggregators. The endpoint supports standard chat-completions workflows without requiring new authentication headers.

Step 3: Test with the /v1/chat/completions Endpoint

Verify your integration by sending a standard POST request to the chat-completions endpoint. Our API supports a context window of 100,000 tokens for both prompt and completion combined. You can generate up to 16,000 tokens per request if you specify max_tokens. The default output limit is 2,048 tokens if you omit this parameter. The endpoint accepts standard parameters like temperature, top_p, and stop. This ensures compatibility with existing prompts and system instructions. You can also enforce JSON mode using response_format for structured data outputs. Testing with a simple text prompt confirms your configuration is correct.

Step 4: Enable Streaming for Real-Time Output

Streaming responses are essential for a smooth user experience in roleplay applications. Our API supports Server-Sent Events (SSE) for token-by-token delivery. You must set the stream parameter to true in your request. The final chunk of the stream includes the total token usage for the request. This allows you to track costs in real-time without additional API calls. Streaming reduces perceived latency by delivering content as soon as it is generated. This feature is critical for maintaining engagement in dynamic conversations. Configure your client to handle SSE streams efficiently.

Step 5: Configure Function Calling

Our API supports function calling with the standard OpenAI tools format. You can define tools and specify tool_choice for automatic or specific execution. This enables complex workflows where the model decides which function to invoke. The API handles tool input parsing and output formatting reliably. You can use this for structured data extraction or automated actions. Function calling works seamlessly with the uncensored model's capabilities. This makes our API suitable for advanced agent architectures that require external tool integration. Ensure your tool definitions follow the standard JSON schema format.

Handling Rate Limits and Errors

You are limited to 300 requests per minute per API key. This allows for high throughput while maintaining server stability. You can have up to 8 concurrent requests active simultaneously. The maximum request body size is 8 MB. If you exceed these limits, you will receive a standard rate limit error. Our pricing is transparent: $0.15 per 1M input tokens and $0.50 per 1M output tokens. Errors and refusals do not incur charges. Credit never expires, so you can top up when convenient. This pay-as-you-go model ensures you only pay for actual usage.

Why This OpenRouter API Alternative is Better for NSFW

Many aggregators apply content filters or route requests through models with varying uncensored levels. Our service serves a single dedicated Dolphin model tuned for unrestricted generation. It does not refuse lawful adult, fictional, or controversial topics. The only hard limit is content involving minors. This consistency is crucial for roleplay and creative writing applications. You avoid the unpredictability of multi-model routers that may switch models mid-session. The dedicated infrastructure ensures that your content preferences are maintained consistently across all requests.

Final Checklist for a Smooth Migration

Before going live, verify your base URL points to our dedicated endpoint. Ensure your API key is active and correctly formatted. Test both streaming and non-streaming responses to confirm stability. Check that your function calling tools are defined correctly. Verify that your token usage tracking aligns with the final SSE chunk. Confirm your payment method is set up for crypto top-ups. Our API accepts USDT on TRC20 or USDC on Base. Start with the trial credit to validate your workflow before committing to larger purchases. This checklist ensures a seamless transition from your previous provider.