AI Tools
Ludwig
Ludwig : prix, capacités, mises à jour et alternatives, lus sur les pages de l’éditeur.

En anglais
Key facts
- Price
- Not stated on the maker’s pages
- Free plan
- Yes
- Platforms
- CLI · API
- API
- Yes
- Open source
- Yes · Apache-2.0 Read on github.com
- Latest version
- v0.17.10
- Last update
- 27 Oct 2026 Read on github.com
Ludwig is a declarative machine-learning framework for researchers and developers. Users define model inputs, outputs, preprocessing, training, and optimization in configuration files, then train models across tabular, text, image, audio, and time-series data.
It supports LLM fine-tuning, AutoML, distributed training, model export, and REST serving. Ludwig is a framework, not a hosted model service; users provide the compute and infrastructure for training and deployment.
- Category
- Developer Tools
- Read from the maker’s pages
- 2026-11-18
Company
Features
- Configure preprocessing, model architecture, training, and metrics in YAML
- Combine tabular, text, image, audio, and time-series features
- Fine-tune language models with methods including LoRA and QLoRA
- Run AutoML and hyperparameter searches with Ray Tune or Optuna
- Scale training across GPUs and distributed Ray clusters
- Serve trained models through REST endpoints and export model files
- Open source under the Apache-2.0 license
Use cases
- Fine-tune a language model on instruction-and-response examples
- Train a classifier using text, image, and tabular product data
- Forecast time-series data with a configured model
- Find a baseline model for a dataset with AutoML
- Deploy a trained model behind a REST endpoint
Capabilities
- Command line — “$ ludwig serve --model_path results/” source
- Choice of models — “Run Llama, Mistral, Qwen, or any HuggingFace model with 4-bit QLoRA on a single consumer GPU.” source
- Self-hosted — “Train locally with CPU or GPU. Fast iteration, no setup.” source
- API — “One command to serve your model as a REST API.” source
- Official SDKs — “train.py — full Python API” source
Latest updates
- Version 0.17.10
Ray workers now return weights as SafeTensors and other results as JSON; the driver no longer unpickles checkpoint data.
- Version 0.17.9
Checkpoint loading now uses PyTorch’s restricted unpickler to prevent code execution from crafted checkpoints.
- v0.17.8: path traversal fix in dataset archive extraction
Fixed path traversal in dataset archive extraction through symlink members.
- Version 0.17.7
Fixed Ray preprocessing tests for Arrow-backed data and row ordering.
- Version 0.17.6
Added preprocessing progress callbacks for pandas, Dask, and Ray backends.
Where it works
CLI · API
Alternatives to Ludwig
- Text Generation Inference
Text Generation Inference is an open-source toolkit for deploying and serving large language models through an API.
- Unsloth
Free, open-source desktop software for running and training AI models locally, with tools for image generation, web search and coding agents.
- Fireworks AI
Fireworks AI hosts, routes, and trains open and custom models for applications, teams, and research.
- openpi
Open-source vision-language-action models and software for training and running robot manipulation policies.
- LoRAX
LoRAX is an open-source inference server for serving multiple fine-tuned LoRA models through shared GPU infrastructure.
- vLLM
vLLM runs and serves large language models on supported hardware, with APIs for applications to send inference requests.
More from Predibase
- LoRAX
LoRAX is an open-source inference server for serving multiple fine-tuned LoRA models through shared GPU infrastructure.
Questions about Ludwig
Is Ludwig free?
Yes. The maker’s pages state that Ludwig can be used for free. Read on the maker’s pages on 18 Nov 2026.
Does Ludwig have an API?
Yes. The maker’s pages state that Ludwig has an API. Read on the maker’s pages on 18 Nov 2026.
Which platforms does Ludwig run on?
Ludwig runs on CLI and API, as its maker’s pages state. Read on the maker’s pages on 18 Nov 2026.
Is Ludwig open source?
Yes. The code of Ludwig is published at github.com/ludwig-ai/ludwig under the Apache-2.0 licence. Read on the maker’s pages on 18 Nov 2026.
Is Ludwig still available?
Yes, as of 18 Nov 2026: its maker’s pages were read that day and carried no notice that Ludwig has ended.
What changed recently in Ludwig?
The latest entry in its maker’s changelog: 27 Oct 2026, “Version 0.17.10”. Before it: 16 Aug 2026, “Version 0.17.9”. Changelog checked on 20 Nov 2026.
Sources
- https://ludwig.ai/ — Official product page describing Ludwig's framework, capabilities, and use cases.
- https://ludwig.ai/latest/user_guide/what_is_ludwig/ — Official documentation covering Ludwig's declarative configuration, multimodal training, optimization, and serving.
- https://github.com/ludwig-ai/ludwig/releases/tag/v0.17.10 — Official repository release page identifying v0.17.10 and its changes.
- https://github.com/ludwig-ai/ludwig — Official repository identifying the Apache-2.0 license.
Not confirmed
- Prices
- Search for Ludwig pricing did not return an official pricing page, so no pricing tiers are listed.
- No official pricing page was found to verify tiers or prices.