SDKs

SeaLink provides an OpenAI-compatible API. The list below covers SDKs we recommend and verify first; other clients that support a custom base URL usually work as well.

Universal pattern

from openai import OpenAI
client = OpenAI(
    base_url="https://test.sealink.io/v1",
    api_key="<your-sealink-key>",
)

SDK	Language	Install	Base URL	API Key	Note
openai-python	Python	pip install openai	base_url	api_key	Official OpenAI Python SDK. Most common path.
openai-node	Node.js / TypeScript	npm install openai	baseURL	apiKey	Official OpenAI Node SDK. Streaming via async iterators.
openai-go	Go	go get github.com/sashabaranov/go-openai	BaseURL	(constructor)	Common community Go client.
openai-rust	Rust	cargo add async-openai	.with_api_base(...)	.with_api_key(...)	async-openai crate; configure it with SeaLink's OpenAI-compatible endpoint.
openai-java	Java / Kotlin	// build.gradle: implementation 'com.openai:openai-java:..'	.baseUrl(...)	.apiKey(...)	Official Java SDK; Kotlin via the same JAR.
openai-dotnet	C# / .NET	dotnet add package OpenAI	ClientOptions.Endpoint	(constructor)	Official OpenAI .NET SDK.
anthropic-python	Python	pip install anthropic	base_url	api_key	Use SeaLink's /anthropic endpoint. Same SDK; switch base_url + api_key.
vercel-ai-sdk	Node / TypeScript	npm install ai @ai-sdk/openai	createOpenAI({ baseURL })	createOpenAI({ apiKey })	Vercel's AI SDK. Use the OpenAI-compatible connection via createOpenAI.
langchain-python	Python	pip install langchain-openai	openai_api_base	openai_api_key	ChatOpenAI(openai_api_base=..., openai_api_key=...)
llamaindex-python	Python	pip install llama-index llama-index-llms-openai	OpenAI(api_base=...)	OpenAI(api_key=...)	LlamaIndex's OpenAI LLM class.

SDKs not listed here may work if they support a custom base_url or baseURL field. If you hit a compatibility issue, email contact@sealink.io.