Ollama-based model manager
Manage local and network LLM providers with a server-oriented architecture prepared for Unix-like systems, VPS machines and dedicated servers.
SKATLAZ · Made in Brazil
Skatlaz Server 2.0
LLM model manager, RAG and GenAI server
Skatlaz Server 2.0 is a pure Python server for managing LLM models based on Ollama, serving as a RAG layer and a corporate GenAI orchestration platform.
RAG for corporate knowledge
Index documents, split content into RAG chunks, organize collections and make company knowledge available to AI agents and chat sessions.
Google GenAI and Gemini integration
Use Google GenAI and Gemini to generate parameterization tasks for AI requirements and create inference packages more quickly.
Agents and categories
Support for Agent categories, AI agents, chat sessions, chat messages, prompt templates and specialized workflows for office automation.
Training and fine tuning
Organize training datasets, fine tune jobs, LLM providers and prompt templates for future model preparation and tailored AI workflows.
Web Diver and web crawling
Run web crawling and web diver tasks to collect public information, RSS feeds and structured content for research and RAG enrichment.
Supported sources
Work with PDF, CSV, XML, TXT, RSS, JSON, HTML, SQL data and database records as sources for corporate AI workflows.
Low-cost distributed computing
Prepared for distributed computing, server farms and Unix-like environments at lower cost, while still being able to run on Windows.
Made in Brazil
Skatlaz is totally Brazilian. Its first official version is focused on practical AI infrastructure, automation and business adoption.
Skatlaz Server modules
Agent categoriesAI agentsChat messagesChat sessionsDocument sourcesFine tune jobsLLM providersPrompt templatesRAG chunksRAG collectionsTraining datasetsWeb diver tasks