Ready
36 blocks · 12 categories

Installable AI blocks that run real models

Full, working experiences that compose the LocalMode UI elements into real apps: chat, knowledge search, vision, audio, photos, and more. Each runs a real model entirely in your browser, with no server and no API key, and nothing downloads until you press Load.

npx shadcn@latest add @localmode/ui/blocks/chat

Install any block the same way - swap chat for any block name.

100%
Private
Zero data leaves device
0ms
Network Latency
After initial model load
Offline
Fully Capable
No internet required
Secure
Local Storage
Encrypted on-device
Zero
Cost & APIs
No cost, no API keys

Chat

1

Chat

Chat with AI models running fully in your browser.

4 providersVision + reasoningAgent mode
~120 MB - 5 GB

Knowledge

4

Semantic Search

Search your text, PDFs, and scans by meaning.

Rerank cross-encodersPDF + OCR ingestCore ⇄ LangChain
~34 MB - 700 MB

Document QA

Ask questions and get answers from your documents.

DistilBERT-SQuADDonut DocVQAConfidence tiers
~34 MB - 800 MB

RAG Chat

Chat with your documents, with cited, grounded answers.

Streaming RAGInline citationsGranite 4.0 350M
~34 MB - 250 MB

Vector Data Manager

Import, export, and inspect your vector database.

4-format importNative exportDrift + reindex
~34 MB

Vision

2

Object Detector

Find and label objects in a photo or webcam frame.

DETR ResNet-50BlazeFace loopWebcam capture
~1 MB - 170 MB

Live Tracker

Track hands, pose, face, and gestures from your webcam.

Hand / Pose / FaceGesture recognitionReal-time FPS
~4 MB - 15 MB

Audio

6

Voice Notes

Record audio and turn it into searchable text notes.

Whisper + MoonshineSynced replayNote search
~34 MB - 237 MB

Live Transcription

Transcribe your microphone live as you speak.

Streaming STTEnergy / Silero VADTurn-taking
~2 MB - 237 MB

Meeting Assistant

Turn a meeting into a summary and action items.

3-step pipelineDistilBART + Granite.txt export
~40 MB - 237 MB

Voice Explorer

Browse and compare 29 natural text-to-speech voices.

29 Kokoro voicesPreviewA/B compare
~86 MB

Audiobook Reader

Turn long text into speech that plays as it streams.

Streaming TTSPause / resumeWAV download
~86 MB

Audio Classifier

Identify the sounds in a recording or audio file.

YAMNet521 categoriesMic or file
~15 MB

Text

1

Language Detector

Detect the language of a text and compare two.

110 languagesAuto-detectText similarity
~7 MB

Device

3

Device Report

See what your device can do for on-device AI.

Capability reportStorage + readinessAdaptive batch
No download

Model Advisor

Get model picks matched to your device and task.

21 tasksCompare + registerRanked models
No download

GGUF Explorer

Browse thousands of downloadable AI models.

160K+ GGUF~4KB Range readChat handoff
No download

Agents

2

Research Agent

Watch an AI agent research a question using tools.

ReAct loopApproval gate3 tools
1.1 GB - 5 GB

Data Extractor

Pull structured data from text into a table and chart.

5 zod templatesRetry/self-correctArtifacts
1.1 GB - 5 GB

Writing Tools

4

Write

Rewrite or improve your draft and review the edits.

AI editsDiff reviewChrome AI + fallback
Chrome AI or ~380 MB

Translate

Translate text between languages, fully offline.

24 offline pairsSwap + copyChrome AI + fallback
Chrome AI or ~80 MB

Summarize

Summarize long text at the length you choose.

Extractive + abstractiveLength presetsCompression stats
Chrome AI or ~120 MB

Complete

Predict the next word and autocomplete your text.

ModernBERT fill-maskTop-5 + applyTransformers.js
~150 MB

Text Insights

4

Sentiment Analyzer

Score text as positive or negative, one or in bulk.

Batch sentimentThroughputWindowed results
~67 MB

Text Classifier

Sort text into your own labels, no training needed.

Zero-shotCustom labelsRanked scores
~25 MB

Model Evaluator

Measure classifier accuracy on labeled data.

P/R/F1 + matrixLabeled datasetsJSON export
~67 MB - 400 MB

Threshold Calibrator

Pick the right similarity cutoff from your data.

Percentile calibrationPreset compareDistribution
~23 MB - 33 MB

Photo

4

Smart Gallery

Auto-sort your photos into categories on device.

Adaptive batchingZero-shot categoriesCLIP ViT-B/32
~350 MB

Image Search

Search your photos by text or by another image.

Text → imageImage → imageOne vector space
~350 MB

Duplicate Finder

Find and remove near-duplicate photos in your library.

Union-findThreshold presetsKeep-first delete
~350 MB

Photo Categorizer

Sort your photos into labels you define.

Editable labelsRe-categorizeFaceted filter
~350 MB

Image Studio

3

Background Remover

Cut out a photo subject to a transparent PNG.

SegFormerTransparent PNGBest-mask alpha
~15 MB

Image Enhancer

Upscale and sharpen a photo at 2x, 4x, or restore.

Swin2SR2x / 4x / RestoreExact upscale
~45 MB / mode

Image Captioner

Generate alt-text captions for your images.

ViT-GPT2Alt-textGallery
~230 MB

Privacy

2

PII Redactor

Find and mask personal details in text, on device.

bert-base-NERRedactionDifferential privacy
~23 MB - 110 MB

Encrypted Vault

A passphrase-locked vault for encrypted notes and files.

AES-GCMAudit logWeb Crypto
No download

Everything runs on your device

What your browser reports it can do for on-device AI. Detected locally, nothing is sent anywhere.

Browser-Reported Capabilities

Detected locally in your browser. Nothing is sent anywhere.

Browser--
Device--
CPU Cores--
Browser Memory--
Browser Storage--
GPU--

Feature Support (hover for details)

WebGPU: not supportedWebNN: not supportedWASM: not supportedSIMD: not supportedThreads: not supportedIndexedDB: not supportedOPFS: not supportedWorkers: not supportedSAB: not supportedSW: not supportedCOI: not supportedChrome AI: not supported