OpenAI-Compatible Chat Completions

Quick Start

Send a chat completion request with a single curl command:

curl -X POST https://api.injectprompt.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{"model": "lite-2.5", "messages": [{"role": "user", "content": "Hello!"}]}'

Supported Models

The API uses model aliases that map to Google Gemini models:

Base URL

https://api.injectprompt.com

Production: https://api.injectprompt.com · Staging: https://staging.api.injectprompt.com

Endpoints

Root — GET /

Simple hello-world endpoint to verify the API is running.

curl https://api.injectprompt.com/

Health — GET /health

Check the health status of the API.

Models — GET /v1/models

Lists all available models with capabilities, context length, and max output tokens.

Chat Completions — POST /v1/chat/completions

OpenAI-compatible chat completions. Requires authentication. Reserves credits before calling the model, then settles actual cost based on token usage.

Billing Balance — GET /v1/billing/balance

Returns the authenticated user's current credit balance. Requires authentication.

Authentication

Protected endpoints require a valid API key in the Authorization header:

Authorization: Bearer sk_live_your_api_key_here

API keys are generated through the InjectPrompt Platform dashboard. Each key is hashed (SHA-256) and stored securely — the plaintext is shown only once on creation.

Protected: /v1/chat/completions, /v1/billing/balance · Public: /, /health, /v1/models

Chat Completions

POST /v1/chat/completions

Parameters

Success Response

{
  "id": "chatcmpl_a1b2c3d4...",
  "object": "chat.completion",
  "created": 1677652288,
  "model": "lite-2.5",
  "choices": [{
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "Hello! How can I help you today?"
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 10,
    "completion_tokens": 20,
    "total_tokens": 30
  }
}

System prompt is injected automatically based on model. Message roles: system, developer, user, assistant. Client-provided system and developer messages are stripped before the provider call.

Credits and Billing

1 credit = $1 USD. Credits are purchased through the InjectPrompt Platform via Stripe.

Pricing (per 1M tokens):

Billing flow: (1) Credits reserved before each request; (2) Actual cost settled after response; (3) If provider fails, reservation released. Use request_id for idempotency — retries with the same ID are not double-charged.

Error Handling

All errors follow the OpenAI error response format:

{
  "error": {
    "message": "Error description",
    "type": "error_type",
    "param": null,
    "code": "error_code"
  }
}

Code Examples

Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    base_url="https://api.injectprompt.com/v1",
    api_key="YOUR_API_KEY"
)

response = client.chat.completions.create(
    model="lite-2.5",
    messages=[{"role": "user", "content": "Hello!"}],
    temperature=0.7,
    max_tokens=150
)
print(response.choices[0].message.content)

JavaScript (OpenAI SDK)

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://api.injectprompt.com/v1',
  apiKey: 'YOUR_API_KEY'
});

const response = await client.chat.completions.create({
  model: 'lite-2.5',
  messages: [{ role: 'user', content: 'Hello!' }],
  temperature: 0.7,
  max_tokens: 150
});

console.log(response.choices[0].message.content);