Integration overview

The recommended sequence is: secure the account → create a dedicated API key → select the protocol and Base URL → fetch a current model → send a minimal request → configure the SDK or client → review usage logs.

Open the Aikey365 console. Account security features may include two-factor authentication, passkeys, and OAuth bindings depending on current configuration.

2. Create a dedicated API key

Use a separate key for each application and environment.

text

local-codex
staging-backend
production-api

This makes revocation, usage review, and access control much easier.

3. Match protocol and Base URL

Scenario	Base URL	Typical endpoint
OpenAI-compatible	https://aikey365.com/v1	/chat/completions, /responses, /models
Anthropic Messages	https://aikey365.com	/v1/messages
Gemini native	https://aikey365.com	/v1beta/models/{model}:generateContent

4. Fetch a current model

bash

curl https://aikey365.com/v1/models \
  -H "Authorization: Bearer $AIKEY365_API_KEY"

Do not guess model identifiers from old tutorials or screenshots.

5. Send a minimal request

bash

curl https://aikey365.com/v1/chat/completions \
  -H "Authorization: Bearer $AIKEY365_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "YOUR_MODEL_ID",
    "messages": [{"role": "user", "content": "Reply with: connected"}]
  }'

Start with plain text and minimal parameters before enabling streaming, tools, vision, or structured output.

Before production

Keep API keys out of frontend code and Git.
Make model IDs, Base URLs, and timeouts configurable.
Use bounded backoff for 429 and transient 5xx responses.
Keep diagnostic metadata without logging full keys or sensitive content.

Integration overview ​

1. Sign in ​

2. Create a dedicated API key ​

3. Match protocol and Base URL ​

4. Fetch a current model ​

5. Send a minimal request ​