Core Features Overview
Explore SparkChat's essential features, from multi-model workspaces to knowledge bases, designed to streamline AI-assisted tasks and automate workflows.
Overview
SparkChat provides powerful tools to enhance your AI interactions. You can set up multi-model workspaces to switch between AI models seamlessly, create custom AI characters and agents for specialized tasks, build a knowledge base using RAG for context-aware responses, and configure workflows with plugins for automation.
These features help you work faster and smarter. Start by exploring the key capabilities below.
Multi-Model Workspaces
Use multiple AI models in one workspace for diverse tasks.
AI Characters & Agents
Build and manage intelligent agents tailored to your needs.
Knowledge Base / RAG
Chat with your documents using retrieval-augmented generation.
Workflows & Plugins
Automate actions with customizable workflows and extensions.
All features are accessible from your SparkChat dashboard at https://app.sparkchat.app.
Multi-Model Workspaces
Set up a multi-model workspace to leverage different AI models like GPT-4, Claude, or Llama in one place. This allows you to compare outputs or assign models to specific tasks.
Create Workspace
Navigate to the workspaces section and click "New Workspace".
Add Models
Select models from the library and configure access with your {API_KEY}.
Switch Models
Use the model selector in chats to toggle between models dynamically.
const sparkchat = new SparkChat({ apiKey: 'YOUR_API_KEY' });
const response = await sparkchat.chat({
model: 'gpt-4',
messages: [{ role: 'user', content: 'Hello' }]
});
from sparkchat import SparkChat
client = SparkChat(api_key='YOUR_API_KEY')
response = client.chat(model='claude-3', messages=[{'role': 'user', 'content': 'Hello'}])
AI Characters and Agents
Create AI characters for role-playing or agents for complex tasks like data analysis. Manage them via the agents dashboard.
Define Role
Set the agent's persona, e.g., "Data Analyst".
Add Tools
Attach plugins like web search or file readers.
Edit, delete, or monitor agent performance from the list view.
Knowledge Base with RAG
Build a personal knowledge base by uploading files. SparkChat uses RAG to retrieve relevant context for accurate responses.
- Go to Knowledge Base > Upload files (PDF, DOCX, etc.).
- SparkChat indexes them automatically.
- Chat with "Summarize this document" to test RAG.
const kb = await sparkchat.knowledgeBase.create({ files: ['doc.pdf'] });
const answer = await sparkchat.chat({
knowledgeBaseId: kb.id,
question: 'What is the main topic?'
});
Limit uploads to <10MB per file for optimal performance.
Workflows and Plugins
Automate repetitive tasks with workflows. Connect plugins for actions like sending emails or updating databases.
{
"name": "Email Alert",
"trigger": "message_received",
"steps": [
{"plugin": "email", "action": "send", "to": "{user.email}"}
]
}
curl -X POST https://api.sparkchat.app/workflows \\
-H "Authorization: Bearer YOUR_TOKEN" \\
-d '{"name": "Email Alert", "trigger": "message_received"}'
Next Steps
Last updated today
Built with Documentation.AI