Developer API

OpenAI-compatible AI API

The SentX API gives your app access to Victoria, the AI model from SentX SI, in the OpenAI Chat Completions request and response format. Point a standard OpenAI client at the SentX base URL and use your SentX key.

Updated

Read the API documentationCreate an API key

At a glance

Base URL · Chat Completions format
https://api.sentx.ai/v3
Model
sentx-3
model name
Input
USD 3.00
per 1M tokens
Output
USD 10.00
per 1M tokens

First request

  1. Sign in, then open Settings → Manage API.
  2. Create a key. Keys start with sk- and are shown once.
  3. Top up your wallet balance (USD 5 to 1,000) and send a request.
Example request (curl)
curl https://api.sentx.ai/v3/chat/completions \
  -H "Authorization: Bearer $SENTX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "sentx-3", "messages": [{"role": "user", "content": "Hello!"}]}'

Compatibility

  • Requests and responses follow the OpenAI Chat Completions format, including streaming.
  • OpenAI SDKs work with the SentX base URL and key; the documentation has Python, JavaScript and curl examples.
  • SentX adds extra fields, such as generated files. Check that your client keeps the fields you need.
  • The legacy /v1 endpoint keeps working. New integrations should use /v3.
  • Compatibility with every client or every OpenAI endpoint is not implied.

Conversation history and memory

The API is stateless: send the full conversation with each request. To give an end user continuity across conversations, pass a stable user value for that person; Victoria keeps per-user memory, and recall is selective.

Pricing and billing

API prices
ItemPrice
Input tokensUSD 3.00 per 1M
Output tokensUSD 10.00 per 1M
Minimum charge1 cent per request
BillingWallet balance

Charges use measured token totals, including context the service adds. Plus does not include API usage; API calls draw from the same wallet balance as paid media.

Rate limits

Requests
150
per rolling 60 s, per key
Streams
34
concurrent, per key
Keys
25
active, per account
New keys
5
per hour, per account

A request over a limit returns HTTP 429 with the code rate_limit_exceeded.

Data and privacy

API requests follow the same Terms of Service and Privacy Policy as the web app. Submissions are not confidential. Keep keys secret and revoke a leaked key at once.

OpenAI is a trademark of OpenAI. SentX SI is not affiliated with OpenAI.

API questions

Can I use the OpenAI SDK?

Yes. Set the base URL to https://api.sentx.ai/v3, use your sk- key and the model name sentx-3.

Does Plus include API usage?

No. API usage is billed per token from your wallet balance.

Does the API remember conversations?

Each request carries its own conversation. Pass a stable user value per end user for memory across conversations; recall is selective.

What are the rate limits?

150 requests per rolling 60 seconds and 34 concurrent streams per key, with up to 25 active keys per account.

Does the /v1 endpoint still work?

Yes. Existing /v1 integrations keep working. Use /v3 for new work.

PricingSign in