methods · software implementation
training framework
13 methods filed at this node or below it, from the sources of 8 models.
In this branch 13
Everything filed at this node or below it, with one collapsible heading per child node.
By model
Which of this branch's methods each model's own documents describe, and how strongly. Under each model: its strongest treatment anywhere in the branch.
| Model | methods |
|---|---|
| NVIDIA-Nemotron-3-Ultra-550B-A55B used | Hugging Face Transformers usedMegatron-LM usedNeMo usedNeMo Gym used— |
| GLM-5.2 optional | Hugging Face Transformers optional— |
| Kolibri-1 core | Declarative training-run configuration coreSavanna core— |
| Laguna-S-2.1 core | Dagster coreExperiments as Code core— |
| MiniMax-M3 optional | Hugging Face Transformers optional— |
| Mistral-Large-4 used | TRL used— |
| NVIDIA-Nemotron-3.5-Super-VL-120B-A12B used | ms-swift usedReinforcement-learning environments in NeMo Gym used— |
| VLA Foundry (Foundry-VLA-1.7B) core | Shared model-agnostic training loop coreTraining with pretrained Hugging Face backbones optional— |