Convert audio between formats (mp3, wav, flac, ogg, aac)

Merge multiple audio files into one

Normalize audio volume levels

Split audio files by silence detection or timestamps

Generate barcodes and QR codes as images

Summarize long text using BART

Classify text into arbitrary categories

Encode and decode Base64 strings

Render templates with batch data (Jinja2, Mustache)

Predict masked tokens in text using BERT

Identify named entities (persons, orgs, locations)

Compact English text embeddings

Generate captions for images using BLIP

ResembleAI expressive TTS

Format source code (Python, JS, Go, Rust, etc.)

Autocomplete code using CodeGen

CogVideo text-to-video 2B

Compress and decompress data (gzip, zstd, brotli)

Parse and evaluate cron expressions

Sesame conversational speech model

Parse and transform CSV data

Monocular depth estimation in-browser via WebGPU

Answer questions given a context passage

Classify text sentiment (positive/negative)

Extract text and metadata from DOCX files

Estimate depth from monocular images

Multilingual text embeddings

Simple echo workload for testing WASM runtime

Detect and blur faces in images for privacy

High-accuracy OCR engine

Fish Audio speech synthesis

Image captioning and visual understanding in-browser via WebGPU

FLUX.1 fast text-to-image

Google Gemma 3 12B — high-quality mid-range

Google Gemma 3 1B — ultra-light, 128K context

Google Gemma 3 27B — large multimodal

Google Gemma 3 4B — multimodal, 128K context

Google Gemma 3n — efficient on-device model, multimodal

Create animated GIFs from images or video clips

GLM-4 32B — strong coding and reasoning

Vision-language OCR with GLM

OpenAI GPT-OSS 20B

Compute cryptographic hashes (SHA256, SHA512, MD5, BLAKE3)

Sanitize HTML to prevent XSS attacks

Convert HTML content to PDF documents

Colorize grayscale images using deep learning

Compress images with quality control (JPEG, PNG, WebP)

Convert images between formats (PNG, JPEG, WebP, AVIF, TIFF)

Compare two images and highlight differences

Generate images from text prompts using Stable Diffusion

Extract EXIF, IPTC, and XMP metadata from images

Resize images with various interpolation methods

IndexTeam TTS model

Small text embeddings from Jina

Generate TypeScript/Python/Go types from JSON samples

Validate, format, and transform JSON (jq-like queries)

Decode, validate, and inspect JWT tokens

Extract keywords from text using embeddings

Lightweight high-quality TTS

On-device text-to-speech on iPhone via CoreML

Text-to-speech synthesis in-browser via WebGPU

Detect the language of input text

Llama 3.1 8B via Exo runtime — single Island or distributed

Meta Llama 3.1 8B Instruct

Real LLM inference using Mistral 7B with llama.cpp

Mock LLM for testing - instant responses

Render Markdown to HTML with syntax highlighting

Generate table of contents from Markdown headings

Convert Markdown to styled PDF documents

Minify HTML, CSS, and JavaScript

Generate text embeddings using MiniLM

Mistral 7B Instruct v0.2 (Q4_K_M)

Mistral Small 3.1 — fast, multimodal, 128K context

Detect faces and generate face embeddings

Apply artistic styles to images

Remove background noise from audio recordings

Convert images/PDFs to Markdown

Extract text from images using optical character recognition

Translate text between languages

NVIDIA Parakeet speech recognition

Extract text, tables, and images from PDF files

Merge multiple PDF files into one document

Add digital signatures to PDF documents

Microsoft Phi-4 Mini 3.8B — reasoning on constrained hardware

Generate QR codes as SVG or PNG

Qwen3.5 0.8B — ultra-light reasoning

On-device LLM inference on iPhone via CoreML

Qwen3.5 27B — top-tier open-source

Qwen3.5 35B MoE (3B active) — efficient large model

Qwen3.5 4B — compact all-rounder

Qwen3.5 9B — best mid-range reasoning

Qwen3 8B — strong all-rounder

Qwen3 automatic speech recognition

Qwen3 Coder Next — 80B MoE, 3B active, best open-source code model

Small Qwen3 text embeddings

Large Qwen3 text embeddings

Qwen3 text-to-speech with custom voices

Multimodal vision-language embeddings

Qwen text-to-image generation

Upscale images 4x using Real-ESRGAN

4x image upscaling in-browser via WebGPU

Test, match, and replace with regular expressions

Capture screenshots of web pages

Text-to-image generation running in-browser via WebGPU

Parse, compare, and validate semantic versions

Generate URL-safe slugs from text

Intelligently crop images to focus on subjects

Tiny GPU-accelerated LLM in-browser via WebGPU

Text-to-image generation (SD 1.5 / SDXL)

Latest SD 3.5 text-to-image

Mock image generation for testing

Optimize and minify SVG files

Render SVG files to PNG images at any resolution

Syntax-highlight source code to HTML

Correct grammar using T5

Paraphrase text using T5

Classic text-to-speech synthesis

Compute unified diffs between texts

On-device LLM inference via llama.cpp

Parse and convert TOML to JSON

Detect toxic content in text

Optical character recognition in-browser via WebGPU

Remove backgrounds from images

Parse, validate, and manipulate URLs

Generate and validate UUIDs (v4, v5, v7)

Microsoft VibeVoice speech recognition

Low-latency streaming TTS

Microsoft VibeVoice TTS

Extract clips from video files by timestamp

Compress video files with quality control

Burn subtitles into video files (SRT, ASS)

Generate thumbnail images from video files

Transcode video between formats (H.264, H.265, VP9, AV1)

Mistral Voxtral speech recognition

Wan-AI text-to-video 1.3B (lightweight)

Wan-AI text-to-video 14B

Wan-AI text-to-video MoE

Add text or image watermarks to images and PDFs

Transcribe audio to text using Whisper

On-device speech-to-text on iPhone via CoreML

Fast large-model transcription

Speech-to-text transcription in-browser via WebGPU

Parse Excel spreadsheets to JSON

Parse and convert YAML to JSON

Detect objects in images using YOLOv8

Segment objects in images using YOLOv8-Seg

Real-time object detection in-browser via WebGPU

Tongyi Z-Image turbo generation