Version 0.1.10-alpha (Oct 12, 2025) - Azure OpenAI & Configuration Improvements

• Azure OpenAI Support: Support for Azure OpenAI including GPT-5 models with automatic parameter handling
• Manual Model Entry: New toggle to bypass automatic model fetching for providers like Azure that don't support /models endpoint
• Improved Error Handling: Fixed 401 errors from external providers no longer logging users out
• Settings Fixes: Mission settings now persist correctly across server restarts with proper priority handling
• UX Improvements: Disabled autocomplete on API key fields to prevent browser autofill issues

MAESTRO is an AI-powered research platform you can host on your own hardware. It's designed to manage complex research tasks from start to finish in a collaborative research environment. Plan your research, let AI agents carry it out, and watch as they generate detailed reports based on your documents and sources from the web.

View Full Documentation

• Quick Start - Get up and running in minutes
• Installation - Platform-specific setup
• Configuration - AI providers and settings
• User Guide - Complete feature guide
• Example Reports - Sample outputs from various models
• Troubleshooting - Common issues and solutions

• Docker and Docker Compose (v2.0+)
• 16GB RAM minimum (32GB recommended)
• 30GB free disk space
• API keys for at least one AI provider

# Clone and setup
git clone https://github.com/murtaza-nasir/maestro.git
cd maestro
./setup-env.sh    # Linux/macOS
# or
.\setup-env.ps1   # Windows PowerShell

# Start services
docker compose up -d

# Monitor startup (takes 5-10 minutes first time)
docker compose logs -f maestro-backend

Access at http://localhost • Default: admin / pass found in .env

For detailed installation instructions, see the Installation Guide.

• CPU Mode: Use docker compose -f docker-compose.cpu.yml up -d
• GPU Support: Automatic detection on Linux/Windows with NVIDIA GPUs
• Network Access: Configure via setup script options

For troubleshooting and advanced configuration, see the documentation.

Azure OpenAI & Configuration Improvements

• Azure OpenAI support including GPT-5 models with automatic parameter handling
• Manual model entry toggle for providers without /models endpoint support
• Fixed 401 errors from external providers no longer logging users out
• Mission settings now persist correctly across server restarts
• Disabled autocomplete on API key fields to prevent browser autofill issues

Stability & Security Update

• Fixed mission pause/resume with proper checkpoint handling
• Replaced passlib with maintained libpass fork
• Resolved Round/Pass counter and activity log persistence issues
• Fixed bcrypt compatibility for authentication

Mission Resilience & Document Intelligence Update

• Intelligent mission resume with complete checkpoint preservation
• arXiv paper fetcher for direct academic paper processing
• Writing phase resume support
• Document reprocessing and re-embedding capabilities
• Fixed progress indicators for accurate research tracking

• Multi-Agent Research System: Planning, Research, Reflection, and Writing agents working in concert
• Advanced RAG Pipeline: Dual BGE-M3 embeddings with PostgreSQL + pgvector
• Document Management: PDF, Word, and Markdown support with semantic search
• Web Integration: Multiple search providers (Tavily, LinkUp, Jina, SearXNG)
• Self-Hosted: Complete control over your data and infrastructure
• Local LLM Support: OpenAI-compatible API for running your own models

This project is dual-licensed:

1. GNU Affero General Public License v3.0 (AGPLv3): MAESTRO is offered under the AGPLv3 as its open-source license.
2. Commercial License: For users or organizations who cannot comply with the AGPLv3, a separate commercial license is available. Please contact the maintainers for more details.

Feedback, bug reports, and feature suggestions are highly valuable. Please feel free to open an Issue on the GitHub repository.