> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cloosphere.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Your First Chat
> Step-by-step guide to selecting an AI model and starting your first conversation
Now that you've signed in, let's start your first AI conversation. This guide covers the basic flow from model selection to using the response.
```mermaid theme={null}
flowchart LR
A["Select Model"] --> B["Type Message"]
B --> C["Send"]
C --> D["Review AI Response"]
D --> E["Use the Response\n(copy, regenerate, etc.)"]
```
***
## Starting a Conversation
Click the **New Chat** button at the top of the sidebar, or use the keyboard shortcut `Ctrl + Shift + O`.
A model selector dropdown and an input box appear in the center of the screen.
Click the **model selector dropdown** at the top of the screen to choose which AI model to use. Clicking the "Select a model" text shows the list of available models.
You can set a frequently used model as your **default**. After picking a model in the dropdown, click the **"Set as default"** button that appears below — it'll be selected automatically in future chats.
Type your question or request in the input box at the bottom of the screen.
Press `Enter` or click the **send button** (arrow icon) at the right of the input box.
Use `Shift + Enter` to insert a line break. Pressing `Enter` alone sends the message.
The AI streams its response in real time, token by token, until generation completes.
Click the **stop button** (square icon) during generation to cancel the response.
***
## Model Selection Guide
Pick the model that best fits your task from the list registered by the admin. Below are typical model characteristics.
| Model Type | Strengths | Recommended For |
| ----------------- | --------------------------------- | ------------------------------------------------ |
| **GPT-4o** | High-performance multimodal model | Complex analysis, image understanding, coding |
| **GPT-4o-mini** | Fast and economical | General chat, simple Q\&A, translation |
| **Claude** | Excellent long-context handling | Document summarization, report writing, analysis |
| **Ollama models** | On-premises execution | Sensitive internal data where security matters |
The list of available models depends on admin settings. If a model you need isn't visible, ask your admin.
***
## Input Options
Beyond plain text, you can pass information to the AI in several ways.
### Attaching Files
Click the **+** button on the left of the input box, or **drag and drop** a file onto the input area.
| File Type | Supported Extensions | Examples |
| ---------------- | -------------------- | ----------------------------------------------------- |
| **Documents** | PDF, DOCX, TXT, MD | Document summarization, content analysis, translation |
| **Spreadsheets** | XLSX, CSV | Data analysis, deriving insights |
| **Images** | PNG, JPG, GIF, WebP | Image analysis, OCR (multimodal model required) |
| **Code** | PY, JS, TS, etc. | Code review, bug analysis, refactoring |
| **Audio** | WAV, MP3, etc. | Auto-transcribed to text on upload (STT) |
| **Data** | JSON, XML, YAML | Structured-data analysis |
Pair file attachments with a specific question to get more accurate answers.
Example: "Summarize chapter 3 of the attached PDF and put the key conclusions in a table."
### Cloud Storage Integration
When configured by the admin, you can pull files directly from **Google Drive**, **OneDrive**, or **SharePoint**. Choose the cloud storage option from the + button menu.
### Commands
Use commands that start with a special symbol in the input box to quickly invoke features.
| Command | Function | Example |
| ------------- | ----------------------------------- | -------------------------------------------------- |
| `@modelname` | Ask a specific model directly | `@gpt-4o review this code` |
| `/promptname` | Invoke a saved prompt template | `/email-draft project delay notice` |
| `#kbname` | Reference a specific Knowledge Base | `#hr-policy What's the leave application process?` |
***
## Using the Response
Action buttons below an AI response provide several follow-up actions.
| Action | Description |
| --------------------- | ----------------------------------------------------------------------------------------------------------- |
| **Edit** | Edit the AI response content directly |
| **Copy** | Copies the entire response to the clipboard |
| **Read Aloud** | Plays the response with TTS (Text-to-Speech) |
| **Good Response** | Response quality feedback (thumbs up) |
| **Bad Response** | Response quality feedback (thumbs down, optional comment) |
| **Continue Response** | Continues generating a truncated response |
| **Regenerate** | Generates a new response for the same prompt. After multiple regenerations you can switch between responses |
| **View Trace** | View tokens used and processing time |
For responses containing code blocks, click the **Copy** button at the top-right of each block to copy just the code.
***
## Continuing the Conversation
Cloosphere maintains conversation context. It remembers previous turns, so follow-up questions flow naturally.
```
[First question] Write Python code that calls a REST API
[Follow-up] Add error handling
[Follow-up] Convert it to an async version
```
As conversations grow long, the influence of earlier messages can fade. When the topic shifts, starting a fresh **New Chat** often produces more accurate answers.
***
## Tips for Effective Prompts
Specific conditions yield more accurate results than vague requests.
| Less effective | More effective |
| ---------------- | ----------------------------------------------------------------------------------------------------------- |
| "Write a report" | "Write a Q1 2024 marketing performance report. About 2 A4 pages, including key KPIs and improvement areas." |
| "Fix this code" | "This Python code throws a TypeError. Find the cause and provide the fixed code." |
Specifying the desired output format reduces post-processing time.
* "Organize as a **table**"
* "Write in **markdown**"
* "Output as **JSON**"
* "Explain step by step with **numbered items**"
Giving the AI a specific role yields more expert-level answers.
* "You're a senior Python developer. Review this code."
* "Analyze this campaign strategy from a marketing expert's perspective."
If you reuse the same role often, register it as an [Agent](/en/workspace/agents) so you don't have to specify the role each time.
For complex work, splitting into steps tends to produce better results than asking for everything at once.
```
Step 1: Find the top 3 trends in the attached data
Step 2: Analyze the cause of each trend
Step 3: Organize the analysis into an executive briefing slide structure
```
***
## Keyboard Shortcuts
| Shortcut | Action |
| ----------------------- | ----------------------- |
| `Enter` | Send message |
| `Shift + Enter` | Line break |
| `Shift + Esc` | Focus chat input |
| `Ctrl + Shift + O` | Open new chat |
| `Ctrl + Shift + S` | Toggle sidebar |
| `Ctrl + Shift + C` | Copy last response |
| `Ctrl + Shift + ;` | Copy last code block |
| `Ctrl + Shift + Enter` | Generate prompt pair |
| `Ctrl + Shift + Delete` | Delete chat |
| `Ctrl + .` | Toggle settings |
| `Ctrl + /` | Show keyboard shortcuts |
***
## Next Steps
Once you've completed your first chat, explore more powerful features.
Learn about the sidebar, search, folders, and the rest of the UI
Organize chats with folders, pinning, and tags
Build task-specific AI assistants to boost productivity