This is the Linux app named 4M whose latest release can be downloaded as ml-4msourcecode.tar.gz. It can be run online in the free hosting provider OnWorks for workstations.
Download and run online this app named 4M with OnWorks for free.
Suivez ces instructions pour exécuter cette application :
- 1. Téléchargé cette application sur votre PC.
- 2. Entrez dans notre gestionnaire de fichiers https://www.onworks.net/myfiles.php?username=XXXXX avec le nom d'utilisateur que vous voulez.
- 3. Téléchargez cette application dans ce gestionnaire de fichiers.
- 4. Démarrez l'émulateur en ligne OnWorks Linux ou Windows en ligne ou l'émulateur en ligne MACOS à partir de ce site Web.
- 5. Depuis le système d'exploitation OnWorks Linux que vous venez de démarrer, accédez à notre gestionnaire de fichiers https://www.onworks.net/myfiles.php?username=XXXXX avec le nom d'utilisateur que vous souhaitez.
- 6. Téléchargez l'application, installez-la et exécutez-la.
CAPTURES D'ÉCRAN
Ad
4M
DESCRIPTION
4M is a training framework for “any-to-any” vision foundation models that uses tokenization and masking to scale across many modalities and tasks. The same model family can classify, segment, detect, caption, and even generate images, with a single interface for both discriminative and generative use. The repository releases code and models for multiple variants (e.g., 4M-7 and 4M-21), emphasizing transfer to unseen tasks and modalities. Training/inference configs and issues discuss things like depth tokenizers, input masks for generation, and CUDA build questions, signaling active research iteration. The design leans into flexibility and steerability, so prompts and masks can shape behavior without bespoke heads per task. In short, 4M provides a unified recipe to pretrain large multimodal models that generalize broadly while remaining practical to fine-tune.
Comment ça marche
- Any-to-any modeling across diverse vision tasks
- Masked modeling with unified tokenization for multiple modalities
- Released model families (e.g., 4M-7, 4M-21) with training/eval code
- Promptable and steerable behavior without task-specific heads
- Transfer to unseen tasks and modalities from a single backbone
- Research-grade configs and examples for reproduction
Langage de programmation
Python
Catégories
This is an application that can also be fetched from https://sourceforge.net/projects/fourm.mirror/. It has been hosted in OnWorks in order to be run online in an easiest way from one of our free Operative Systems.
