WebGPU.Studio
AI models running entirely in your browser, powered by WebGPU
Try These First
Our most polished experiences, ready to go
Speech to Text
Real-time transcription powered by OpenAI Whisper. Record from your mic or upload audio files.
Chat
Conversational AI with streaming responses and thinking support. Runs local LLMs directly in your browser.
Object Detection
Point your camera or upload a photo to detect and label objects in real-time with bounding boxes.
Speech & Audio
Speech to Text
Record or upload audio for real-time transcription with Whisper
Text to Speech
Convert text to natural-sounding speech with LFM2.5 Audio or SpeechT5
LFM Audio Studio
Unified ASR, TTS, and near-real-time interleaved voice conversation
Music Generation
Generate music from text prompts with Meta's MusicGen
Vision & Image
Background Removal
Instantly remove image backgrounds with RMBG
Object Detection
Real-time object detection with bounding boxes and labels
Depth Estimation
Generate depth maps from 2D images with Depth Anything V2
Image Segmentation
Click to segment objects with Meta's SAM3 — upload or use your camera
Vision Chat
Upload images and ask questions about them with a vision-language model