Studio-Chat

Studio-Chat is an intelligent conversational tool provided by the ZenMux platform. It supports advanced features such as multi-model chat, PK mode, Artifact generation, image generation, and more. This document will help you quickly get up to speed with Studio-Chat’s core capabilities.

Core Features

Start a New Chat

1. Visit Studio-Chat
2. Enter your question in the input box at the bottom of the page
3. Press Enter or click the send button to submit your message
4. Wait for the AI model to generate a response

Select a Model

In the model selector at the top of the chat area, you can:

• Click the model name to open the model list
• Search for and select the AI model you need
• Different models support different capabilities (text, images, audio, etc.)

PK Mode (Multi-Model Comparison)

PK mode lets you chat with multiple AI models at the same time, making it easy to compare response quality across models.

Enable PK Mode

1. In the model selection area, click the PK icon (➕) to add a new chat window
2. You can add up to 10 models for side-by-side comparison
3. Each window can select a different model independently

Sync in PK Mode

PK mode supports synchronized sending:

Icon	Status	Description
🔛 On	Sync On	When you send a message, all synced models receive it
🔚 Off	Sync Off	This model window runs independently and won’t be synced

How to use:

1. Click the sync toggle icon in the model selection area
2. When enabled, a message sent from any input box will be broadcast to all models with sync enabled
3. When disabled, that model window requires separate input and sending

Remove a PK Window

Click the × button in the model selection area to remove the corresponding chat window.

Artifacts

Artifacts is a powerful content generation and preview feature that supports generating interactive code, charts, documents, and more.

Supported Artifact Types

Type	MIME Type	Description
Markdown	text/markdown	Rich text documents
HTML	text/html	Interactive web content
SVG	image/svg+xml	Vector graphics
Mermaid	application/vnd.mermaid	Flowcharts, sequence diagrams, etc.

Use Artifacts

1. In the input area, click Skills and select Artifact Mode (💡 icon), or click the Artifacts button in the input area
2. Send your request, for example:
   • "Help me create a project flowchart"
   • "Generate an interactive HTML page"
   • "Draw a logo in SVG"
3. After the AI generates the content, the Artifact preview panel opens automatically

Artifact Panel Features

• Content / Preview toggle: View source code or preview the rendered output
• Copy code: One-click copy of the generated content
• Download file: Save the content as a file in the corresponding format
• Fullscreen mode: Expand the preview area

Web Search

Web Search allows the AI to search the internet for the latest information while answering your question.

Models That Support Web Search

⚠️ Note: Not all models support Web Search.

How to tell:

• In the Skills menu, if Web Search is clickable (not grayed out), the current model supports it
• If it’s grayed out/disabled, the current model does not support it

Common models that support Web Search:

• Models that support the web_search_options parameter (e.g., some OpenAI and Google models)

Use Web Search

1. Click Skills
2. Enable Web Search (🌐 icon)
3. Send your question
4. The AI will search the web and answer based on the latest information

When enabled, responses will include cited sources (grounding chunks) to help you verify the information.

Image Generation & Editing

Image Generation Mode

Some models support image generation:

1. Click Skills
2. Select Image Mode (🖼️ icon)
3. Describe the image you want to generate
4. The AI will generate the corresponding image

Special settings for Google Imagen models:

When using Google’s image generation models, you can configure:

• Image Aspect Ratio: Aspect ratio (1:1, 16:9, 9:16, etc.)
• Image Resolution: Output resolution

Image Editing Mode

Models that support image output can also perform image editing:

Mode	Description
Basic Edit	Basic editing to modify image content
Cutout	Cutout: remove/replace the background
Upscale	Image upscaling (not available yet)

Steps:

1. Upload an image
2. In Skills, select the corresponding editing mode
3. Describe the edit you want
4. The AI will return the edited image

Voice Input & Output

Voice Input

Models that support voice input can accept audio messages:

1. Click the microphone icon in the input box to start recording
2. Once recording is complete, it will upload automatically
3. When you send the message, the audio will be sent to the AI along with it

Voice Output

Some models support spoken replies:

1. Enable Audio Mode (🎵 icon) in Skills
2. Send a message
3. The AI’s response will include audio content that you can play directly

API Call Example

Click the </> icon in the model selection area to view and copy an API call example for the model.

Advanced Parameters

Click the ⚙️ Settings icon in the model selection area to configure advanced model parameters.

Configurable Parameters

Parameter	Description	Range
Provider	Select the model provider	Model-dependent
System Prompt	System prompt that defines the AI’s role and behavior	Text
Enable Reasoning	Enable reasoning mode (chain-of-thought)	On/Off
Max Reasoning Tokens	Max tokens for reasoning	0-100000
Max Completion Tokens	Max output tokens	0-262144
Temperature	Controls randomness of the output	0.0-2.0
Top P	Nucleus sampling	0.0-1.0
Top K	Only consider the top K most likely tokens	0-100
Frequency Penalty	Frequency penalty	-2.0 to 2.0
Presence Penalty	Presence penalty	-2.0 to 2.0
Repetition Penalty	Repetition penalty	0.0-2.0
Min P	Minimum probability threshold	0.0-1.0
Top A	Adaptive sampling parameter	0.0-1.0

Smart Routing Preferences

If you are using an auto-routing model, you can set routing preferences:

• Performance: Prefer the highest-performance provider
• Price: Prefer the lowest-cost provider
• Balanced: Balance performance and price

Reasoning Mode Notes

• Required: For some models (e.g., the o1 series), reasoning mode is mandatory
• Optional: Some models allow you to enable reasoning mode selectively
• Not supported: Some models do not support reasoning mode

Billing Mode (PAYGo vs Subscription)

ZenMux offers two billing modes, allowing you to choose flexibly based on your usage.

Subscription Mode

Item	Description
💰 Billing	Monthly/annual subscription
✅ Best for	High-frequency users
📋 Coverage	Models included in your plan
🔸 Icon	Yellow subscription icon

Pay As You Go Mode

Item	Description
💰 Billing	Pay based on actual usage
✅ Best for	Low-frequency usage or models not in your plan
📋 Coverage	All models
🔸 Icon	Blue API icon

Switch Billing Modes

1. Click the billing icon (yellow/blue) next to the model selection area
2. In the dropdown menu, choose:
   • Subscription Mode - Use subscription quota
   • Pay As You Go Mode - Use your account balance

Notes

• If the selected model is not included in your subscription plan, the system will automatically switch to Pay As You Go mode
• Some advanced models may require sufficient account balance to use
• The system will automatically detect and prompt you to top up or upgrade your subscription

File Upload

Studio-Chat supports uploading multiple file types.

Supported File Types

Specific file types depend on model support. If an option is grayed out/disabled, the current model does not support it.

Type	Formats	Description
Images	jpg, png, gif, webp, etc.	Image understanding & analysis
Documents	pdf	PDF document analysis
Text	txt, md, json, yaml, etc.	Text file analysis
Video	mp4, etc.	Video understanding
Audio	wav, mp3, etc.	Speech-to-text

Upload Methods

1. Click to upload: Click the + button on the left side of the input box, then choose a file type to upload
2. Drag and drop: Drag files directly into the input area
3. Paste to upload: Copy an image and paste it into the input box

File Preview

Uploaded files are shown as thumbnails; click to preview:

• Images: view full-size
• PDF: view PDF content online
• Text: view file contents

Session Management

Session List

The left sidebar shows your chat history, supporting:

• Create a new session: Click the "New Chat" button
• Switch sessions: Click any historical session
• Delete sessions: Right-click a session item or click the delete button
• Rename sessions: Double-click the session name to edit

Session Grouping

Sessions are automatically grouped by time:

• Today
• Yesterday
• Last 7 Days
• Last 30 Days
• Older

Session Saving

• Each turn is saved automatically
• Session titles default to the content of your first message
• You can return at any time to continue previous conversations

Shortcuts

Action	Shortcut/Method
Send message	Enter
New line	Shift + Enter
Stop generating	Click the Stop button
Copy reply	Click the copy icon below a message
Regenerate	Click the refresh icon below a message

FAQ

Q: Why are some features unavailable?

A: Different models support different features. Make sure the selected model supports the feature you need.

Q: How many models does PK mode support?

A: Up to 10 models can be compared simultaneously.

Q: How do I know whether a model supports Web Search?

A: If the Web Search option in the Skills menu is clickable, it’s supported.

Q: Can Subscription and Pay As You Go be used at the same time?

A: Yes. Each chat window can select its billing mode independently.

Q: Is there a file size limit for uploads?

A: Yes. The specific limit depends on the file type and what the current model supports.

Get Help

Contact Us

If you encounter any issues during use, or have suggestions or feedback, feel free to contact us via:

• Official website: https://zenmux.ai
• Support email: [email protected]
• Business inquiries: [email protected]
• Twitter: @ZenMuxAI
• Discord community: http://discord.gg/vHZZzj84Bm

For more contact options and details, please visit our Contact Us page.