Izanagi - AI Unified Interface

Control Panel

ComfyUI

Offline

Port: 8188

Status: Not running

Ollama

Offline

Port: 11434

Status: Not running

Installed Models: -

TTS

Offline

Status: Not running

Available Voices: -

Code Execution

Offline

Status: Not running

Provider: -

Model: -

Logs

Settings

Network Access Allow other computers to connect

Code Execution Mode Only show models compatible with code execution

Ollama Model Path (Optional)

Agent Model The model to use for code execution (provider-dependent)

Hugging Face API Key Required for Hugging Face models

OpenAI API Key Required for OpenAI models

ComfyUI

Offline

ComfyUI is not running

Start ComfyUI from the Control Panel to access it here

Ollama

Offline

New Chat

Welcome to Ollama Chat

Select a model and start chatting!

Text-To-Speech

Input Text

Audio Output

Voice:

Quality:

Generate audio from your text to listen here

Code Execution Offline

Not available

Model:

Izanagi Webstudio

Live Preview

Chat memory helps the AI remember your conversation history and preferences across sessions. The interaction history tracks topics you've discussed to provide context for future conversations.

Preferred Name

How you prefer the AI to address you

Addressing Style

How formal or casual the AI should be

Tone Preference

The overall tone of AI responses

Response Length

How detailed or brief responses should be

Explanation Depth

How technical explanations should be

Use Emojis

Whether the AI should use emojis

Interaction History

Summary of your interactions with the AI, tracking topics and sentiment over time

Advanced Performance Settings

Configure model performance parameters. Changes require model reload to take effect.

GPU Layers: -1 = Auto (use all GPU), 0 = CPU Only. Higher values use more GPU memory but are faster.

Context Size (tokens): Larger context allows for more code but uses more memory.

Max Output Tokens:

Temperature: 0.1 Controls randomness. Lower = more deterministic code.

Top P: 0.95 Controls diversity of generation. Lower = more focused output.

Unlimited Generation (No Auto-Stop) ⚠️ When enabled, generation will NEVER stop automatically and will use full max_tokens. This removes all early stopping conditions for maximum length output.

Note: Changes require model reload to take effect. The model will be automatically reloaded when you change these settings.

Control Panel

ComfyUI

Ollama

TTS

Code Execution

Logs

Settings

ComfyUI

ComfyUI is not running

Ollama

Welcome to Ollama Chat

Text-To-Speech

Input Text

Audio Output

Izanagi Webstudio

Live Preview

Model Settings

System Prompt

Generation Parameters

Agent Provider & Model

Ollama Models

Rename Chat

Clone Voice

Voice Settings

Voice Parameters

Chat Memory Settings

Izanagi Webstudio Models

Advanced Performance Settings