Settings
Global Defaults
these defaults are used when provider-specific values are not set
API Type
OpenRouter
OpenAI
Custom
Base URL
openai compatible endpoint with /chat/completions
API Key
stored only in your browser
Refresh model list
Generation Defaults
applies to the active model request
Temperature
0.70
Max tokens
2048
Evaluation
Validation mode
Check AI responses with a second model call
OFF
Providers
each provider can have its own url key and default model
Add Provider
Session
Export session JSON
Clear saved key & settings
destructive action you must type CLEAR
System Prompt
{{sources}}
{{language}}
{{today}}
{{notebook_name}}
Library
Prompt Library
Add Provider
Provider Name
Base URL
API Key
Default Model
Untitled notebook
Configure API
SOURCES
(0)
Add
Add your first source to start chatting
↑ Drop PDF, DOCX, TXT, MD, CSV, PY, TS and other text files here
Fetch
~0k / 16k tokens
SESSIONS
New Chat
What would you like to explore?
Add a source on the left, or try one of these:
Summarize
What are the main ideas in this document?
Key facts
List the most important facts and figures.
Clarify
Explain this topic in simple terms with examples.
Deep dive
Explain the methodology in detail.
Details
Sources
Hybrid
Model
~0 tokens
·
$0.0000
System
Model
Studio
Validation ON
~0 tokens
Preview
Notes
Studio
✕
Click a citation
to view the source
Notes
auto-saved
▦
Flashcards
Anki export
?
Quiz
MCQ + open
◉
Mindmap
Concept map
▤
Report
Full summary
◌
Podcast
Dialogue script
i
FAQ
Q&A list
←
Quiz
Count
Type
Multiple Choice (MCQ)
True / False
Open-ended
Focus (optional)
Generate
📋 Copy
+ Notebook
↺ Regen
No saved snippets yet.
Click
+ Notebook
to save.