Tool to Build & Manage LLMs on your own infrastructure
Greenative Studio
Bridging generative AI and various databases and storages
Greenative MCP Servers
Extending Generative AI
greenative-skills
Data platform to improve Data and AI usability
Greenative Platform
Greenative enables natural language-based data processing (ETL) using generative AI. Through the Greenative Platform, it facilitates data processing across various databases and storage systems within a private environment. It incorporates industry-standard generative AI extension technologies such as the Model Context Protocol (MCP) and skills.
Additionally, it allows users to fine-tune open generative AI models using their own data.
The Greenative Platform (formerly Avgidea Data Platform) can be used as a data hub to connect various systems, databases, and storage solutions.
It streamlines data integration and business processes—which can often become complex within an organization—while maintaining the independence of each system.
Event Processing Added to Greenative Platform
2026.9.24
Greenative Platform now lets you run functions in an event-driven way. When an object is uploaded to Avgidea Storage, the upload event triggers an Event function.
For example, you can capture an event when a CSV file is uploaded to Storage, retrieve the file using an Event function, and execute business logic.
Click here for more details.
Local LLM Coding to Greenative Studio is released
2026.9.3
We announced a major update to Greenative Studio, its private AI workbench for Mac.
The new Coding release adds an Anthropic-compatible API gateway that lets agentic coding tools — including Claude Code — run against large language models executing locally on the user's Mac, and introduces support for NVIDIA's newly released Nemotron 3.5 Lightning model.
A free trial is also available.
Click here for more details.
Greenative Studio for macOS is released
2026.7.7
Greenative Studio is a macOS app for running open LLMs locally on a Mac. It does not require complex setups like Python or Homebrew, and allows you to use the models without sending prompts or business data to cloud-based LLMs.
The main features of Greenative Studio are as follows.
→ Local Fine-Tuning
You can fine-tune models on your Mac to suit your company’s business data.
→ MCP Integration
You can integrate models with files and various tools.
→ Support for Open Models Suitable for Commercial Use
You can focus on using models that are licensed for commercial use.
→ Fixed-Rate Pricing
You can use the service without worrying about pay-as-you-go token-based billing.
We believe that local LLMs will become an increasingly valuable option in scenarios where data privacy, cost predictability, and control over the environment are critical.
A free trial is also available.
Click here for more details.
greenative-skills is now publicly available on GitHub.
2026.2.24
We have released the "greenative-skills" extension for Anthropic's AI coding agent "Claude Code" on GitHub.
This feature enables direct operation of the Greenative Platform from Claude Code, allowing natural language-based data processing and ETL (Extract, Transform, Load) operations to be performed on various data sources.
GitHub Repository: https://github.com/greenative-ai/skills
Blog
Introducing Avgidea KVS
2026.2.5
We’re excited to release Avgidea KVS, a new key-value store component on the Avgidea Data Platform.
What is it ? A lightweight storage solution that lets you store and retrieve JSON data with simple put and get operations.
Why use it ?
-
Simple API — no complex setup
-
Multiple stores for organizing your data
-
Seamless integration with Avgidea Data Platform
-
Perfect for configurations, caching, and lightweight data persistence
Whether you're building applications that need quick data access or looking for a straightforward way to manage key-value pairs, Avgidea KVS has you covered.
📘 Learn more: https://doc.avgidea.io/kvs
Rolled out new updates for Greenative
2026.1.16
🤗 Hugging Face integration
🧠 Fine-tuning support for safetensors models
🦙 Support for Llama 3
⚡ Quantization to GGUF (GPT-Generated Unified Format)
Greenative
