Overview
GPU-powered inference for image, video, OCR, TTS, and ASR models.
On this page
Overview
The Inference Platform provides API access to production-ready ML models for image generation, video generation, OCR, text-to-speech, and speech recognition. Models run on GPU infrastructure and are billed per-inference in credits.
Available capabilities
- Text-to-Image — generate images from text prompts
- Image-to-Image — edit and transform existing images
- Text-to-Video — generate video from text descriptions
- OCR — extract text from images and documents
- Text-to-Speech — generate natural speech from text
- Speech Recognition — transcribe audio to text
How it works
- 1Browse the model catalog for available models
- 2Call the invoke API with a model slug and input
- 3Platform runs inference on GPU infrastructure
- 4Results are returned synchronously or via webhook
- 5Credits are deducted based on model pricing