Guides
Framework and Tool Integrations
Connect APIQIK to AI frameworks and developer tools with a custom OpenAI Base URL, then verify compatibility with minimal requests.
Tools that accept a custom OpenAI Base URL and API key can usually connect to APIQIK. Third-party frameworks may send extra parameters automatically, so verify compatibility before production use.
Common configuration
| Setting | Value | Notes |
|---|---|---|
| API type | OpenAI-compatible | Prefer Chat Completions rather than the unconfirmed Responses API. |
| Base URL | https://api.apiqik.com/v1 | Some tools require /v1; others append the version path automatically. |
| API key | APIQIK API key | Use environment variables or secure secret storage, never public configuration. |
| Model | Exact ID from the model catalog | Do not assume a framework's default model is available on APIQIK. |
Common integrations
LangChain
Use the OpenAI Chat model adapter and test non-streaming, streaming, and tool calls.
LlamaIndex
Configure an OpenAI-compatible LLM. Confirm embeddings separately rather than assuming the same integration applies.
Vercel AI SDK
Set a custom Base URL on the OpenAI provider and validate SSE forwarding.
Developer tools
Compatibility with tools such as Cursor and Claude Code depends on their custom-provider and endpoint requirements.
Verification order
- Send a non-streaming text-only request to verify the URL, key, and model.
- Enable streaming and verify incremental chunks and terminal events.
- If tools are required, validate argument assembly, tool messages, and multi-turn loops.
- Check for parameters automatically added by the framework that APIQIK or the target model does not support.
- Pin validated framework versions and run regression tests after upgrades.