⚡ edgeFlow.js

Lightweight Browser ML Inference Framework

🖥️
Runtime
📊
Memory
📦
Load ONNX Model
Download and initialize real ONNX models from Hugging Face
✂️
Interactive Segmentation
Click to segment objects - powered by SAM (~14MB)
Loading SAM model...
0%
Loading SAM model... Please wait.
💬
AI Chat
TinyLlama 1.1B running in your browser (~714MB)

TinyLlama is a 1.1B parameter language model.

Download size: ~714MB (may take several minutes)

Click "Load Model" to start
🧮
Tensor Operations
Create, manipulate, and compute tensors
📝
Text Classification
Sentiment analysis with loaded model
🔍
Feature Extraction
Extract embeddings from text
📦
Model Quantization
Compress tensors to int8/float16
🔬
Tensor Debugger
Inspect tensor statistics & distribution
⚡
Benchmark
Measure operation performance
📋
Task Scheduler
Priority-based task scheduling
💾
Memory Management
Allocate, track, and release tensors
⚡
Concurrent Execution
Run multiple inferences in parallel
📊
Performance Monitor
Real-time performance metrics and visualization