Skip to main content

Make Your First Model Request

Choose a maintained path below, verify each result in the Console, and open the detailed task page when you need exact fields, commands, or troubleshooting.

Open Chat Completions
API Overview

The model API is OpenAI-compatible where documented. Use the environment gateway, send your API key as a Bearer token, choose a model ID from Console, and start with the endpoint for its modality.

Docs 2.0 · API contract 2.0.9

What You Can Do

  • Identify the current environment gateway URL.
  • Authenticate without embedding a key in source code.
  • Choose a model ID that matches the endpoint modality.
  • Send a minimal request and read its response.
  • Use trace and rate-limit headers when troubleshooting.
  • Find the matching TypeScript or Python next step.

Who This Is For

Application developers integrating model inference · Teams moving an OpenAI-compatible client to Smart Studio · Developers generating chat, images, or video without Console automation

Choose Your Path

01Recommended

Chat Completions

Text, reasoning, and multimodal chat requests.

  1. Set Base URL
  2. Export API key
  3. Choose model
  4. POST /v1/chat/completions
  5. Read response

Result: You receive an authenticated chat completion and can trace the request.

Open Chat API
02

Image Generation

Text-to-image or image-guided generation.

  1. Choose image model
  2. Build input
  3. POST /v1/images/generations
  4. Save task ID
  5. Poll result

Result: An asynchronous image task is created with a task ID.

Open Image API
03

Video Generation

Text, image, keyframe, or reference-to-video generation.

  1. Choose video model
  2. Select input mode
  3. POST /v1/videos/generations
  4. Save task ID
  5. Poll result

Result: An asynchronous video task is created with a task ID.

Open Video API

Before You Start

  • An API key created in Tenant Console → API Keys.
  • A callable model ID copied from Models or the matching Lab.
  • Available credits or billing for the selected model.

Core Journey

1Set Base URL2Export API key3Choose model4POST /v1/chat/completions5Read response

Expected Result

You can send one authenticated request to the current gateway, identify its trace information, and continue to the full contract for the selected endpoint.

Recommended Next Step

Open Chat Completions

Current Environment​

  • Gateway Base URL: https://gateway.example.com/model-gateway
  • Authentication: Authorization: Bearer $SMART_STUDIO_API_KEY
  • Model IDs: copy the callable identifier from Console → Models or the matching Lab.
export SMART_STUDIO_BASE_URL="https://your-environment-gateway"
export SMART_STUDIO_API_KEY="your-api-key"

curl "$SMART_STUDIO_BASE_URL/v1/chat/completions" \
  -H "Authorization: Bearer $SMART_STUDIO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"your-model-id","messages":[{"role":"user","content":"Hello"}]}'

Use the Base URL shown above for SMART_STUDIO_BASE_URL. Keep the real API key outside shell history, source files, and logs.