Features

Everything CogmemUI brings to your Ai workflow — one interface, every capability.

Core Capabilities

💬

Multi-Model Chat

14 Ai models across 6 providers. Switch between Claude, GPT, Gemini, Grok, DeepSeek, and Mistral in one click.

🧠

Persistent Memory

Powered by CogmemAi with the Intelligence Engine — a self-improving memory system. Semantic search finds memories by meaning. Related knowledge auto-links into a knowledge graph. Contradictions are flagged automatically. Ask a question and get one synthesized answer from all your memories. Auto-Skills go further: your corrections become behavioral rules that teach your Ai how to work with you. Choose cloud (recommended — full intelligence), local (zero-config, no account), or hybrid (both).

🛠️

Full Tool Use

11 server-side tools: web search, image generation, file operations, code execution, and more — all used automatically.

🎤

Voice Chat

Continuous listening, auto-send after silence, Ai responses read aloud, and tap-to-interrupt support.

🎨

Canvas Editor

Side-by-side code and content editing with live preview, version history, and Ai-powered actions.

📄

Document RAG

Upload PDFs, DOCX, and text files. Documents are automatically chunked, indexed, and referenced in conversation.

Advanced Features

⚔️

Model Arena

Blind side-by-side comparison of 2–3 models. Vote on the best response to reveal model names and track performance.

🔄

Workflow Builder

Visual drag-and-drop editor for multi-step Ai workflows. Nodes execute with real-time progress streaming.

🚀

App Deploy

Turn any HTML, JS, or SVG creation into a live web app with one click. Includes public/private toggle and embed snippet.

🤖

Agent Mode

Autonomous task execution with a live control panel. The Ai creates a step-by-step plan, executes it using tools, and streams progress in real time. Pause, resume, or stop at any point. Risk-gated approvals keep you in control of high-impact actions. Supports parallel sub-agents for multi-part tasks and background execution that keeps running even if you disconnect.

🔎

Deep Research

Multi-step research mode that explores topics from every angle with web searches, source citations, and structured reports.

📋

Conversation Templates

Save and reuse conversation starters with {{variable}} placeholders. Template Runner generates dynamic input forms.

📊

Analytics Dashboard

Track usage, costs, model performance, and conversation insights with interactive charts and KPI cards.

💰

Budget Controls

Set daily and monthly spending limits. Auto-downgrade to cheaper models when approaching your budget cap.

🤖

Auto Router

Select "Auto" and CogmemUI picks the best model for each message based on task type and your budget.

🧩

Extensions

Install custom slash commands, tools, UI panels, and event hooks. Build extensions with the CogmemUI API.

🧠

Shared Team Memory

Your team’s collective knowledge becomes shared Ai memory. Decisions, protocols, best practices, and institutional knowledge are saved by any team member and proactively surfaced to everyone when relevant. Every memory includes author attribution so you know who contributed it. Learn more →

👥

Team Roles & Access

Three clear roles: Owner (full control), Admin (manage members and conversations), and Member (participate and access team knowledge). Create multiple teams — one per project, department, or client. Requires CogmemAi Teams tier.

💬

Conversation Sharing

Share any conversation with your team in one click. Shared conversations appear in every member’s sidebar with a team badge. Private conversations stay completely private — only what you share becomes visible.

🟢

Real-Time Presence

See who’s online with colored presence avatars. See who’s viewing the same conversation. See real-time typing indicators when a teammate is composing a message. Your team feels connected even when working remotely.

🔒

Scope Isolation

Personal memories and team memories are completely separate. Toggle between scopes with one click in the Memory Panel. Your personal knowledge stays yours. Team knowledge belongs to everyone. Zero bleed between scopes.

💻

Local Companion

Route chat through your Claude Code subscription with npm install -g cogmemui-local. Access local files, CogmemAi memory tools, and your own compute.

🔗

MCP Server Connectivity

Connect any MCP-compatible tool server via the open Model Context Protocol standard. Go to Settings > MCP Servers to add servers, auto-discover their tools, and use them in chat automatically. Each user manages their own server connections independently. Give your Ai access to databases, APIs, custom tools, and more. Setup guide →

🗺️

Mind Maps

Visualize ideas, topics, and conversation threads as interactive spatial mind maps you can drag and expand.

🎭

Personas

Create custom Ai personas with different system prompts and personalities. Switch between them instantly.

📅

Scheduled Messages

Set up messages that send automatically on a schedule. Daily briefings, regular checks, and timed reminders.

📝

Plan Mode

Have the Ai outline a structured plan before executing. Review and adjust the approach before work begins.

💬

@ Mentions

Reference other conversations, memories, or models inline. Context is automatically injected for the Ai.

💡

Think Panel

Watch the Ai's extended reasoning process in real-time with a collapsible side panel and token counter.

📌

Pinned Tools

Quick-access toolbar for your 10 most-used tools. One click to trigger web search, image gen, memory, and more.

📤

Sharing & Export

Share conversations via link, export as JSON, or generate formatted PDFs. Import conversations from JSON files.

🔀

Git Integration

View diffs, branches, and commit history. The Ai can help manage code changes directly from the chat.

35+ Slash Commands

CogmemUI includes over 35 slash commands for instant access to every feature, plus 7 Ai prefix commands that modify how the Ai responds. Type / to see them all, or check the complete reference in the User Guide.

Tip: Use /do to launch Agent Mode for autonomous multi-step tasks, /research for deep investigations, and /compare to send the same prompt to all your models at once.

Supported Ai Models

Provider Models Strengths
Anthropic Opus 4.6, Sonnet 4.6, Haiku 4.5 Coding, analysis, long context
OpenAI GPT-4o, GPT-4o Mini, o1 General purpose, image generation
Google Gemini 2.0 Pro, Gemini 2.0 Flash Multimodal, Google ecosystem
xAI Grok 3, Grok 3 Mini Real-time knowledge, creative tasks
DeepSeek DeepSeek V3, DeepSeek R1 Reasoning, math, cost-effective
Mistral Mistral Large, Mistral Small Multilingual, efficient, coding
Tip: Use Model Arena to compare any 2–3 models on the same prompt. It's the fastest way to find the best model for your use case.

Themes

CogmemUI supports four visual themes to match your preference:

  • Light — Clean white interface, easy on the eyes in bright environments
  • Dark — Deep blacks with indigo accents, great for low-light use
  • Auto — Follows your operating system's light/dark preference
  • Terminal — Retro green-on-black aesthetic for the hacker at heart

Switch themes with the /theme slash command or in Settings > Appearance.

Keyboard Shortcuts

Ctrl + K
Open the command palette
Ctrl + N
Start a new conversation
Ctrl + F
Search within the current conversation
Ctrl + /
Show all keyboard shortcuts
Enter
Send message
Shift + Enter
New line (without sending)
Ctrl + Shift + C
Copy last Ai response
Escape
Close modals, panels, or menus

Works Everywhere

CogmemUI is a Progressive Web App. Install it on your phone, tablet, or desktop for a native-app experience with offline support. Works in Chrome, Firefox, Edge, and Safari.