PersonalAI

Local-First Engine

User Profile

Local mode

Ollama —

Provider —

PersonalAI

What can I help you with?

v1.0.0 · — Latency · — Tokens Connecting · Log: DEBUG

Live Events STREAMING

GPU UTIL —

System Performance

Tokens/sec

—

Context

—

Latency —

Total tokens —

Explorer

src/core

assistant.ts .ts

context.ts .ts

logger.ts .ts

src/providers

ollama.ts .ts

anthropic.ts .ts

openai.ts .ts

src/tools

registry.ts .ts

calculator.ts .ts

Coding Assistant

Switch to Coder profile for optimized code assistance using qwen2.5-coder.

Quick Actions

Recent Memories

Knowledge Graph

Enter a query above to search the web

Results appear here. Memories shown on the left.

Total Memories

—

stored entries

Search Engine

LIKE search

Avg / Entry

—

characters

Architecture

SQLite

WAL mode

Disk Usage

—

KB on disk

Vector Topology

Vault Operations

Retention Policies

Auto-archive

Auto-save

Max entries ∞

Model Providers

Hardware Context

RAM —

Loading…

CPU—

CPU LOAD—

THERMAL STATUS—

SWAP LATENCY—

Primary model—

Provider—

Tool use—

Task Routing Table

Task Type	Model	Fallback Provider	Reason
tools	qwen2.5:14b	ollama	Native tool-use support
coding	qwen2.5:14b	ollama	Code generation
reasoning	qwen2.5:14b	ollama	Strong logical reasoning
chat	gemma3:12b	ollama	Conversational, low latency
longcontext	gemma3:12b	ollama	128k context window
quick	gemma3:12b	ollama	Fast responses

Registered Tools

Loading…