Overview Dashboard

Real-time status of WhatsApp RAG collector and Hermes Agent on Vast.ai

Vast.ai 24/7 PM2 Service: ONLINE
Total Messages Indexed
1,482
+142 today
RAG Context Chunks
296
Indexed into Vector Memory
Active Conversations
38
100% Silent Logger Mode
Vast.ai PM2 Status
ONLINE
Auto-reboot enabled (@reboot cron)

Live WhatsApp RAG Stream

Real-Time Ingestion

Vast.ai Infrastructure

vast-hermes
Cloud Host: ssh6.vast.ai (18.234.176.169)
PM2 Service: whatsapp-rag-bot (PID: 190514)
Hermes Gateway: Running (PID: 188852)
Auto-Restart: Active via crontab (@reboot)
Data Store: rag_knowledge_base.json (Pure JS)

Quick Commands

Search WhatsApp RAG Knowledge Base

296 Chunks Indexed

OpenAI & Gemini Fine-Tuning Exporter

The WhatsApp RAG Collector automatically extracts Q&A pairs from conversation threads and converts them into standardized JSONL format for fine-tuning models like Llama 3, Mistral, GPT-4o, or Gemini.

148 Clean Q&A Training Pairs
100% Privacy Masked & Cleaned

Preview (whatsapp_dataset.jsonl)

Loading training dataset...

Holographic Hermes RAG Playground

Connected to Vast.ai Model & WhatsApp RAG Store
Interactive Inference
Holographic Hermes AI
Hello! I am Holographic Hermes, connected directly to your WhatsApp RAG Knowledge Base on Vast.ai. Ask me any question based on your indexed WhatsApp data!
Just now
Host Instance
ssh6.vast.ai
Port 11106
Memory Engine
Pure JS JSON
Zero C++ Segfaults
Cron Reboot Rule
@reboot
PM2 Resurrect Active

Vast.ai Terminal Controls & Instructions

Use these standard CLI commands in your terminal to manage your Vast.ai instance directly:

1. View Live Background PM2 Logs
ssh vast-hermes "pm2 logs whatsapp-rag-bot"
2. Export Latest Fine-Tuning JSONL File
ssh vast-hermes "cd ~/whatsapp-bot && npm run export"
3. Launch Interactive Holographic Hermes CLI Query
ssh vast-hermes -t "cd ~/whatsapp-bot && npm run ask-hermes"