Skip to content

plexe

Model Serving & MLOps

A tool that builds machine learning models from natural‑language descriptions of the desired task and dataset.

Track releases GitHub Website
Python Latest v1.4.4 · 2mo ago Security brief →

Features

  • Multi‑agent workflow automatically analyzes data, selects metrics, searches for models, evaluates them, and packages a deployable artifact
  • Supports multiple ML frameworks (XGBoost, CatBoost, LightGBM, Keras, PyTorch) with configurable model type selection
  • Provides Docker images with all dependencies pre‑installed and a Makefile for common workflows
  • YAML configuration for customizing LLM routing, search parameters, Spark settings, and more
  • Multi‑provider LLM support via LiteLLM (OpenAI, Anthropic, Ollama, etc.)

Recent releases

View all 15 releases →
v1.4.4 New feature
Notable features
  • Support for explicit split inputs
  • Ability to override neural network (NN) epoch count
Full changelog

What's Changed

  • chore: ignore .claude and .codex directories by @marcellodebernardi in https://github.com/plexe-ai/plexe/pull/188
  • feat: support explicit split inputs and nn epoch overrides by @marcellodebernardi in https://github.com/plexe-ai/plexe/pull/189

Full Changelog: https://github.com/plexe-ai/plexe/compare/v1.4.3...v1.4.4

View release on GitHub
v1.4.3 New feature
Notable features
  • Add probability‑aware metric support across search and evaluation
Full changelog

What's Changed

  • feat: add probability-aware metric support across search and evaluation by @marcellodebernardi in https://github.com/plexe-ai/plexe/pull/187

Full Changelog: https://github.com/plexe-ai/plexe/compare/v1.4.2...v1.4.3

View release on GitHub
v1.4.2 Bug fix

Fixed calibration of evaluator verdict rubric and baseline context.

Full changelog

What's Changed

  • fix: calibrate evaluator verdict rubric and baseline context by @marcellodebernardi in https://github.com/plexe-ai/plexe/pull/186

Full Changelog: https://github.com/plexe-ai/plexe/compare/v1.4.1...v1.4.2

View release on GitHub
v1.4.1 New feature
Notable features
  • Train/val gap tracking
  • Bootstrap insights
Full changelog

What's Changed

  • feat: add train/val gap tracking and bootstrap insights by @marcellodebernardi in https://github.com/plexe-ai/plexe/pull/185

Full Changelog: https://github.com/plexe-ai/plexe/compare/v1.4.0...v1.4.1

View release on GitHub
v1.4.0 New feature
Notable features
  • Added GPU‑aware Keras and PyTorch training runtime
Full changelog

What's Changed

  • feat: add gpu-aware keras/pytorch training runtime by @marcellodebernardi in https://github.com/plexe-ai/plexe/pull/183

Full Changelog: https://github.com/plexe-ai/plexe/compare/v1.3.6...v1.4.0

View release on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

About

Stars
2,581
Forks
256
Languages
Python Makefile Dockerfile
View on GitHub Homepage

Install & Platforms

Install via
pip
Platforms
linux macos windows arm64

Community & Support

Discord

Similar tools

plex-rewind
Plonk
Reaparr
medialytics
pasta

Beta — feedback welcome: [email protected]