vIvai
📎 Files Attached
📚 RAG Collections
Sessions
C
U
R
A
I
T
D
N
Chat Sessions
Current Session
Default Session
Saved Sessions
Chat Indexing
What this does: Indexes your chat history for semantic search, allowing the AI to find and include relevant previous conversations in your prompts.
How it works: Similar to how documents are processed for RAG, chat messages are converted to embeddings and stored in the vector database for fast retrieval.
Chat History Index:
Checking...
Indexed Messages:
0
Last Indexed:
Never
Indexing Configuration
Include relevant chat history in prompts using semantic search
History Weight
0.3
Weight of chat history vs document context (0 = documents only, 1 =
history only)
Max History Results
3
Maximum number of chat history messages to include
Automatically index new chat messages for semantic search (triggers
Sync New for each message)
RAG Configuration
RAG System Status
RAG Server:
Checking...
Vector Database:
Checking...
Embedding Model:
Current Collection:
None
Collection Status:
Unknown
Vector Collections
0 collections
Embedding Configuration
Include relevant context from vector database in prompts
Similarity Threshold
0.7
Minimum similarity score for including context (0 = include all, 1 =
exact match only)
Max Context Results
5
Maximum number of context chunks to include in prompt
Context Window Size
512
Maximum tokens per context chunk
Document Management
Current Folder:
Root
Total Items
0
RAG Indexed
0
Selected
0
Current Model
384D
Query Testing
🔍 Image Search & Management
0.1
🖼️ Thumbnail Management
×
Search Results
0 results
Search time: 0ms
Chat History Count
2
Number of previous chat messages to include in each session (0 =
none, 10 = max)
File Management
Central File Storage
Prompt Templates
Debug Config
Generation Parameters
Temperature
0.7
Controls randomness: 0 = deterministic, 1 = creative
Max Tokens
2048
Maximum number of tokens to generate
Top-k
40
Number of highest probability tokens to consider
Top-p
0.95
Samples from top 95% probability mass, smooth diversity
Repeat Penalty
1.0
No penalty; allows normal repetition behavior
Min-p
0.05
Minimum probability for token selection
Programming Notes
🤖 Adaptive Human Intelligence (AHI)
System Status
AHI Server:
Checking...
Database:
Checking...
Models:
Checking...
📷 Image Analysis
Select a model to see details
Drag and drop images here or click to select
Supports: JPG, PNG, BMP, TIFF
🔍 Analysis Results
No analysis results yet. Upload an image to get started.
👥 Identified Humans
No humans identified yet. Process some images to see results.
📁 Document Management
Current Folder:
Root
Total Items
0
AHI Analyzed
0
Selected
0
Current Model
MediaPipe
📋 Recognition Sessions
No recognition sessions yet.