What does OpenAI-compatible access mean here?
Compatible access uses the relevant request structure, authentication and response format. Start with a Chat Completions text request: model selects the model, messages describes the conversation and choices contains the ordinary response.
Chat Completions, Responses, image generation and other APIs are separate endpoints. A successful text request does not establish support for every OpenAI API or parameter. Check Tikway's documentation, the selected model and the endpoint your application actually calls.
- Choose an available model that supports the target protocol and endpoint.
- Check the SDK version, timeouts and retry behavior.
- Validate tools, structured output and multimodal features individually.
Set Python base_url or Node.js baseURL correctly
For the OpenAI SDK Chat Completions example, the base address is https://api.tikway.ai/v1. The SDK appends /chat/completions. Putting the full endpoint in the base URL can produce a duplicated path.
Configure authentication with your own Tikway key and copy the model request ID from the catalog. Some clients append /v1; others expect a complete URL. Inspect the resulting request path before deciding what to enter.
- Python uses base_url; Node.js uses baseURL.
- Read the key from server-side environment variables, not public frontend code.
- For 401, check authentication. For 404 or path errors, check URL composition.
Check GPT pricing and streaming against your workload
Compare model versions, context windows, output limits and supported parameters. Estimate ordinary text input and output separately. When models or routing options change, use the current catalog and account rules rather than an old quote.
Parse streaming responses as events, preserving data across network chunks and UTF-8 boundaries. Check in-stream errors and termination signals. Retries can add calls and costs; validate a short request before moving to production workloads.
Endpoints & protocol
| Setting | Value or verification step |
|---|---|
| SDK base URL | https://api.tikway.ai/v1 |
| Complete Chat Completions endpoint | POST https://api.tikway.ai/v1/chat/completions |
| Authentication & model | Authorization: Bearer your Tikway key · model: request ID from the catalog |
Check prices before estimating a budget
Base text cost = input tokens × input rate + output tokens × output rate
Base catalog prices are in USD per token. Multiply by 1,000,000 to compare a per-million-token quote. The calculator estimates ordinary text input and output only. Check cache, image, video and other charges separately; account billing rules and request records determine the final charge.
Integration & troubleshooting resources
Frequently asked questions
Do all GPT models support the same endpoints?
Do not infer support from the GPT name alone. Check endpoints, parameters and capabilities in the selected model's catalog entry and documentation, then test the request.
Why does a custom base URL still return 404?
Check whether your client appends /v1 or /chat/completions and avoid repeating those segments. Also verify that the model supports the requested endpoint.
Can I use my OpenAI credits on Tikway?
Tikway is an independent service with its own keys and account billing. OpenAI credits and ChatGPT subscriptions do not combine with Tikway billing.