Sprelay API Docs

SPRELAY DOCS / AI API PLATFORM

Sprelay API Docs

One gateway for multiple AI models

Start with your account, API key, and model group, then follow step-by-step client and API guides with examples you can copy and verify.

Start from step one Open dashboard

BASE URLhttps://sprelaytoken.com/v1

POST/v1/chat/completions200 OK

curl https://sprelaytoken.com/v1/chat/completions \
  -H "Authorization: Bearer $SPRELAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.6-sol",
    "messages": [
      {"role": "user", "content": "Reply only: --------Connection successful--------"}
    ]
  }'

RESPONSE--------Connection successful--------

OpenAIResponsesClaudeGemini

FIRST-TIME SETUP

Connect in four steps

Set up your account, key, model, and first verified request in order.

Prepare your account

Create an account, sign in, and confirm that the dashboard is available.

Create an API key

Choose a model group and set spending and access limits for the key.

Choose a model

Check live model availability, protocols, groups, and pricing in the catalog.

Make the first request

Verify the URL, key, and model with a minimal request before configuring an app.

CLIENT SETUP

Find the guide for your tool

Choose the client you already use, or continue to a protocol guide for development.

View all client guides

CC SwitchImport Claude, Codex, and Gemini settings Cherry StudioDesktop chat and knowledge-base client Claude CodeAgentic coding in the terminal Codex CLIConnect Codex with the Responses protocol CursorConfigure a custom OpenAI base URL Other clientsFor OpenAI-compatible tools

DEVELOPER INTEGRATION

Choose the protocol used by your project

One Sprelay key can be used with multiple supported SDK and request formats.

OPENAI COMPATIBLE

Chat Completions

The broadest-compatible chat endpoint for existing SDKs and clients.

View guide OPENAI RESPONSES

Responses

A unified input and output API for modern agents and tool workflows.

View guide ANTHROPIC NATIVE

Claude Messages

Use Anthropic message structures and native request headers.

View guide GEMINI NATIVE

Gemini generateContent

Use Gemini-native paths, content structures, and authentication.

COMMON ISSUES

Troubleshoot by symptom

01401 / 403 authentication errorsCheck key format, status, quota, and access restrictions.02429 too many requestsDistinguish rate limits, concurrency limits, and upstream congestion.03Client cannot connectCheck base URLs, automatically appended paths, and proxy settings.04Streaming output stopsCheck client support, buffering, timeouts, and event parsing.

Never expose an API key in screenshots, chat logs, or frontend code. Read the API key security guide