Getting Started
Playground
Interactive browser environment for evaluating models directly before writing code.
The Playground lives at /dashboard/playground behind account authentication. It provides a dedicated testing surface for running prompts, tuning sampling parameters, and evaluating outputs across six modalities.
An account is required to access the Playground. New accounts receive free credit upon registration, and an initial API key is generated automatically in your account dashboard.
API Parity
The Playground is an interactive client running directly against the same public REST API described in the rest of this documentation. Every parameter, prompt structure, and response format configured in the interface behaves identically when called from your code.
Playground Tabs
Chat
The Chat tab sends conversational messages against any text-generation or multimodal model, with configurable controls for system prompts, temperature, max tokens, and top-p. Models with extended reasoning capabilities display their chain of thought in a separate collapsible block alongside the stream. See the Chat Completions documentation to integrate conversational endpoints.
Images
The Images tab supports both text-to-image generation and image editing workflows. You can generate images from descriptive prompts or upload source images for transformation, with response formats supporting hosted image URLs (url) or base64-encoded data (b64_json). See the Image Generation documentation for API payload options.
Text to Speech
The Text to Speech tab converts written text into natural spoken audio with in-browser playback. Each supported voice model includes a list of selectable preset voices to evaluate vocal tone and characteristics before deployment. See the Text to Speech documentation to synthesize speech from code.
Speech to Text
The Speech to Text tab transcribes audio recordings into text using automated speech recognition models. You can record voice directly using your browser microphone or upload pre-recorded audio files in MP3, WAV, M4A, or WEBM formats up to a stated maximum file size of 25 MB. See the Speech Recognition documentation for transcription parameters.
Video
The Video tab enables text-to-video generation and video editing. Because video rendering is computationally intensive, requests run synchronously and may hold the HTTP connection open for several minutes until video generation completes. See the Text to Video documentation for request structures.
OCR
The OCR tab extracts text and structural content from document images. Upload an image along with an extraction instruction, which is formatted and sent through the vision content format on /v1/chat/completions. See the Vision & OCR documentation for multimodal formatting details.