OpenAI · Anthropic · Gemini compatible

XPNCORE API documentation

All you need is a base URL, an API key and a model ID. OpenAI, Anthropic and Gemini SDKs all work.

Open the setup wizard Playground User guide

Base URL

https://api.xpncore.com/v1

Add the endpoint path after it, for example /chat/completions.

At a glance

Don't hard-code model IDs. Call GET /v1/models with your API key to get the models you can use.

Bearer authentication

Just one Authorization header.

Clear limits

Limits per API key, plan and wallet balance.

Streaming (SSE)

Sent in parts, ending with [DONE].

Authentication

Send your API key as a Bearer token on every request.

HTTP header

Authorization: Bearer YOUR_API_KEY

Endpoints

Every endpoint XPNCORE supports.

POST
/v1/chat/completions
Create an OpenAI-compatible chat completion.
POST
/v1/combo
Quote, fan out, and synthesize one request across multiple models.
GET
/v1/models
List models available to the current API key.
GET
/v1/models/*model
Retrieve one model available to the current API key.
POST
/v1/embeddings
Create vector embeddings for text input.
POST
/v1/messages
Create a message using the Anthropic-compatible contract.
POST
/v1/messages/count_tokens
Estimate the input tokens of an Anthropic Messages request without sending it or billing it.
POST
/v1/completions
Create a legacy text completion.
POST
/v1/images/generations
Generate images from a text prompt.
POST
/v1/images/edits
Edit an image from an image and prompt.
POST
/v1/images/variations
Create variations of an uploaded image.
POST
/v1/audio/speech
Generate speech audio from text.
POST
/v1/audio/transcriptions
Transcribe audio into text.
POST
/v1/audio/translations
Translate audio into English text.
GET
/v1/realtime
docs.endpoint.realtime.summary
POST
/v1/rerank
Rerank documents against a query.
POST
/v1/moderations
Classify potentially unsafe input.
POST
/v1/search
Search the web for a query through a search-capable model.
POST
/v1/web/fetch
Read a web page's content from its URL through a fetch-capable model.
GET
/v1beta/models
List models through the Gemini-compatible contract.
POST
/v1beta/models/*path
Generate content through the Gemini-compatible contract.
POST
/v1/responses
Create a response with the OpenAI Responses contract.
POST
/v1/responses/compact
Compact a Responses conversation context.
POST
/v1/videos
Make a video from a prompt or images, then poll and download it.

Code examples

Pick an endpoint and a language; every example reads the API key from an environment variable.

Example endpoint

POST /v1/chat/completions

curl https://api.xpncore.com/v1/chat/completions \
  -H "Authorization: Bearer $XPNCORE_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"Hello"}]}'

Try it live

Requests go straight from your browser; the API key stays only in this tab.

API key

Endpoint

Model

JSON body

208 / 32,768 bytes

Response

The response will appear here.

Headers

No headers yet.

Responses and errors

What you receive, with and without streaming.

200

OpenAI-compatible JSON with choices, usage, and model.

200 SSE

Each data event carries choices[].delta.content; [DONE] marks the end.

4xx / 5xx

Errors come with a clear message: authentication, permission, allowance, invalid input, or a model that is temporarily failing.

200 · application/json

{
  "id": "chatcmpl_...",
  "object": "chat.completion",
  "model": "provider/model",
  "choices": [{
    "index": 0,
    "message": { "role": "assistant", "content": "Hello" },
    "finish_reason": "stop"
  }],
  "usage": { "prompt_tokens": 12, "completion_tokens": 4, "total_tokens": 16 }
}

200 · text/event-stream

data: {"object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"Hel"}}]}

data: {"object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"lo"}}]}

data: [DONE]

Limits

Limits that apply to every request.

Request size: Up to 2 MiB (Anthropic Messages and Responses: 32 MiB)
Timeout: Depends on the model; use stream=true for long answers
Streaming: stream=false JSON · stream=true text/event-stream · [DONE]
Finding models: GET /v1/models per API key
Rate limit: Per API key and active plan

Ready to send a request?

The setup wizard helps you create an API key; the Playground lets you try requests right in the browser.

Start setup Playground