AI

Use the AI API in your apps and automations with a dedicated API key, text and image inputs, streaming, and structured responses.

Link: AI

Login/Register

1. Open Login.
2. Sign in with Apple, Google, or an email code. New users are registered automatically.
3. The same account manages your AI services, balance, and billing documents.

Top up balance

1. Open Payment Methods and add funds using an available method.
2. Keep enough balance for the monthly price shown before purchase and for future renewals.

Buy the service

1. Open AI.
2. Enter a name for your app or integration, then press Continue.
3. Review the monthly price and confirm creation. Open the AI order in Profile to view its API key and requests-per-minute limit.

Connect your application

Set AI_API_KEY securely to the generated key. Keep it on your server; do not embed it in public website code or mobile app bundles.

For clients with a configurable API base URL, use https://yoosolutions.com/api/v1/ai and model ai. The provider and model assigned to your order handle the request; feature availability depends on that model.

curl --request POST 'https://yoosolutions.com/api/v1/ai/chat/completions' \
  --header "Authorization: Bearer ${AI_API_KEY}" \
  --header 'Content-Type: application/json' \
  --data '{"model":"ai","messages":[{"role":"user","content":"Summarize the progress of this project."}]}'

Read the reply from choices[0].message.content in the JSON response.

Supported requests

• Send text in messages. User messages can also contain image_url parts with base64 JPEG, PNG, or WebP data; remote image URLs are rejected.
• Use response_format with text, json_object, or json_schema. JSON-schema output requires strict: true.
• Set stream: true for streaming. To include usage, also set stream_options.include_usage: true.
• Control output length with max_completion_tokens. If supplied, store must be false; other undocumented request fields are rejected.
• Requests are stateless. Include the conversation history needed for each request in messages.

Limits and billing

• Requests are limited to 20 MiB, 120 seconds, and 16384 output tokens, with 4096 output tokens by default.
• The requests-per-minute limit shown in your order uses a rolling 60-second window and continues across key resets.
• Renewals are charged from your balance. If renewal is unpaid, API access is blocked until successful payment. Billing history, invoices, and receipts are available in Profile.

Manage your service

• Edit changes the service name without changing the key, model, or limit.
• Reset generates a new API key and immediately revokes the old key. Update your application to use the new key; resetting does not clear the rate limit.
• Remove revokes API access and stops future billing for the service.

Troubleshooting

• 400: check the request fields, model ai, and message format.
• 401: check the bearer key and update it after a reset.
• 402: check your balance and renewal payment in Profile.
• 429: wait for the number of seconds specified by the Retry-After header before retrying.
• 503: the assigned model or provider is unavailable. Try again later or contact Support.