Smart Studio documentation site is now live — covering the full product with interactive API reference, user guides, and hands-on tutorials.

API Reference​

Complete OpenAPI-based interactive API documentation, generated from gateway specifications with live playground support.

Chat​

• Create Chat Completion (POST /v1/chat/completions) — Multi-model chat with streaming, function calling, and vision support
• Create Response (POST /v1/responses) — Responses API for structured output

Embeddings​

• Create Embeddings (POST /v1/embeddings) — Text embedding generation for search and RAG

Rerank​

• Rerank Documents (POST /v1/rerank) — Document relevance reranking with token usage tracking

Images​

• Submit Image Generation Task (POST /v1/images/generations) — Text-to-image and image-to-image generation (async)

Videos​

• Submit Video Generation Task (POST /v1/videos/generations) — Text-to-video and image-to-video generation (async)

Audio​

• Text-to-Speech Synthesis — TTS with configurable voice presets and language hints
• Submit ASR Transcription Task — Speech-to-text transcription (async)

Tasks​

• Get Async Task Result (GET /v1/tasks/{task_id}) — Poll image/video/audio task status and results

Apps​

• Submit App Inference (Async) (POST /v1/apps/{code}/inference) — Run custom app workflows asynchronously
• Submit App Inference (Sync) (POST /v1/apps/{code}/inference/sync) — Synchronous app inference
• Get App Task Result (GET /v1/apps/tasks/{task_id}) — Poll app task results

User Guides​

Getting Started​

• API Key Management — Create, rotate, and manage API keys with permission controls; API key details page with usage metrics (Total API Calls, Tokens, Model Usage Breakdown)
• Gateway Management — Configure and monitor your dedicated API gateway
• Authorization — Role-based access control and team permissions
• Router Quickstart — Step-by-step guide to make your first API call

Model Gallery​

• Explore Models — Browse open-source and commercial models with real-time pricing
• Commercial Model Cards — Pricing display (input/output cost), code samples (Python, TypeScript, Java, Shell)
• Model Selection Guide — How to choose the right model for your use case

Model Lab​

• Chat Playground — Interactive chat with LLMs, supporting multi-turn conversation and parameter tuning
• Image Generation — Text-to-image generation with prompt engineering
• Video Generation — Text/image-to-video generation with task tracking
• Chat History — View and replay past conversations including image and video outputs

Datasets​

• Create Dataset — Upload and organize training datasets
• AI Dataset Preparation — Automated data preparation with AI-assisted labeling
• Manage Datasets — Version control, filtering, and dataset operations

Deployments​

• Create Deployment — Deploy fine-tuned models to dedicated endpoints
• Deploy HuggingFace Models — Import and deploy models from HuggingFace
• Manage Deployments — Monitor, scale, and manage deployment lifecycle

Fine-Tuning​

• SFT (LLM/VLM) — Supervised fine-tuning for language and vision models
• DPO (LLM/VLM) — Direct Preference Optimization alignment training
• Knowledge Distillation — Transfer knowledge from large to small models
• Data Distillation — Generate synthetic training data from teacher models

Usage & Billing​

• Usage Dashboard — Usage charts (Requests, Tokens, Latency), API key filter, data export
• Billing — Cost tracking and billing details

Platform Management​

• Provider Keys — Configure third-party model provider credentials
• Cluster Management — GPU cluster provisioning and management
• Dedicated Gateway — Private gateway deployment and configuration
• Access Management — Team member roles and permissions

API Marketplace​

• API Market — Discover and subscribe to third-party API services