Chat
ARCA CORE

Your private intelligence.

Arca v0.3.5 adds a real model runtime, memory and improved document reasoning.

A
Arca Core v0.3.5 is ready. Open Arca Core and start the Browser Brain once. After the model loads, Chat and Documents can reason without cloud credits.CORE SYSTEM
Core routing: browser brain -> QVAC -> optional cloud

Documents

Upload PDF, DOCX, TXT or Markdown. Arca retrieves relevant passages and gives them to the brain.

Tap to choose files

Documents are parsed locally in this browser.

Ask your documents

No active document.

Arca Core

This is the brain layer: model runtime, context, memory, retrieval and engine routing.

CORE

Browser Brain

Open-source LLM running on your device with WebGPU. Model cache: IndexedDB.

Checking WebGPU support...

Local model

Core routing

...WEBGPU
OFFBRAIN
0MEMORY TURNS
0DOCUMENTS
Reasoning

The model interprets meaning instead of matching exact words.

Core
Memory

Recent conversation turns are retained and supplied to the model.

Core
Document RAG

Relevant chunks are retrieved before the model answers.

Core
Tool Router

Prepared for WordPress, server and XHM actions later.

Next

Voice

Speak to Arca, then send the transcription through the same Core.

Tap to speak

Uses browser speech recognition when available.

Connections

Arca Core can switch model providers without changing the rest of Arca.

BB Browser Brain

Runs an open-source model directly in this browser.

Checking
QL QVAC Local

OpenAI-compatible local endpoint for installed QVAC.

Optional
CL Cloud AI

Optional fallback provider for harder tasks.

Not configured
DF Document RAG

Local parsing, chunking and retrieval.

Ready
WP WordPress

Diagnostics and approved actions through Core tools.

Next
WS Website / Server

Health, logs and diagnostics through Core tools.

Next

Settings

Configure optional QVAC and cloud fallbacks. The Browser Brain requires no API key.

The first Browser Brain start downloads an open-source model and can take time/data. After caching, later starts are much faster. Model files stay in browser storage.

QVAC/local endpoint

QVAC model alias

Cloud API base URL

Cloud model

Cloud API key (session only)

Document context characters

Memory turns