Building AI projects often means switching between multiple tools—for datasets, models, annotation, benchmarking, training, and deployment.
I wanted a simpler workflow, so I built MLForge.
MLForge is a free, local-first platform that brings the entire computer vision workflow into one place.
🔍 Discover 20,000+ AI models and datasets from leading providers
🏷️ Built-in annotation studio
📊 Benchmark models and hardware
⚡ Run inference with an intuitive dashboard
🚀 Train computer vision models
📦 Deploy models from a single interface
💻 Supports Windows, Linux, and macOS
Instead of managing multiple tools and workflows, MLForge provides a unified workspace for ML engineers, researchers, students, and AI developers.
Whether you're starting a new project or deploying a production model, everything is available in one platform.
🌐 Website: https://mlforge.in
🎥 Demo: https://youtu.be/V0YBemyTkG8
I'm actively developing MLForge and would love your feedback. If there's a feature you'd like to see or something that could improve your workflow, let me know.
Bringing dataset discovery, annotation, benchmarking, and deployment into one workflow addresses a lot of the friction in early ML projects.
The interesting challenge is balancing simplicity for new users while still giving experienced ML engineers enough control over their workflows.