Quick facts
- Best for
- Automodel - Native Training Library for LLMs and VLMs
- Pricing
- Free
- Editor rating
- 4.5 / 5
- Community saves
- 0
About Automodel
Automodel, published under the NVIDIA-NeMo GitHub organization, is an Apache-2.0 PyTorch-native distributed training library for LLMs and VLMs. Its README links documentation, ready-to-use recipes, examples, model coverage, and performance material, and notes support work for model families such as Qwen, Nemotron, Kimi, Gemma, Mistral, DeepSeek, and VLM checkpoints.
Pros
- PyTorch-native distributed training workflows for LLMs and VLMs
- Out-of-the-box Hugging Face model support
- Ready-to-use recipe files for fine-tuning and training tasks
- Documentation for model coverage and performance summaries
- Examples for LLM and VLM fine-tuning paths
- Support activity across current model families including Qwen, Nemotron, Kimi, Gemma, Mistral, and DeepSeek topics
- Apache-2.0 source distribution for review and extension
Cons
Pricing
Open source
$0
- • Source code available on GitHub
- • No listed SaaS subscription required
- • Users pay only for their own compute, API, or hosting costs