5. Endpoints and SDK Examples

Version: August 18, 2026

7. Chat completions

Endpoint: POST https://api.logicc.cloud/v1/chat/completions

Tool calling is available when supported by the selected model. Tool inputs must always be valid JSON objects. Compatibility for parameters such as reasoning_effort, structured output, vision, and file handling varies by model.

8. Python SDK example

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["LOGICC_API_KEY"],
    base_url="https://api.logicc.cloud/v1",
)

response = client.chat.completions.create(
    model="gpt-5-mini",
    messages=[
        {"role": "user", "content": "Write a concise project status update."}
    ],
)

print(response.choices[0].message.content)

9. JavaScript SDK example

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.LOGICC_API_KEY,
  baseURL: "https://api.logicc.cloud/v1",
});

const response = await client.chat.completions.create({
  model: "gpt-5-mini",
  messages: [
    { role: "user", content: "Create a concise meeting summary." },
  ],
});

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

10. Streaming

Set stream to true to receive server-sent events when the selected model supports streaming. Clients must handle partial chunks, connection interruptions, and a final completion event.

11. Embeddings

Endpoint: POST https://api.logicc.cloud/v1/embeddings

curl -X POST "https://api.logicc.cloud/v1/embeddings" \
  -H "Authorization: Bearer $LOGICC_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "text-embedding-3-small",
    "input": ["First document", "Second document"]
  }'

The embedding model must be enabled for the key and returned by GET /v1/models.

12. Audio transcription

Endpoint: POST https://api.logicc.cloud/v1/audio/transcriptions

curl "https://api.logicc.cloud/v1/audio/transcriptions" \
  -H "Authorization: Bearer $LOGICC_API_KEY" \
  -F "model=azure/gpt-4o-transcribe" \
  -F "[email protected]"

The request uses multipart/form-data. Common formats include MP3, MP4, MPEG, MPGA, M4A, WAV, and WebM. Depending on the enabled model, optional parameters can include language, prompt, and response_format.

14. Response structure

Provider-specific metadata can be returned in additional fields, including routing information, the resolved model, latency, and the original error response.