# NVIDIA FastGen

## Instructions for AI Agents

- For clean Markdown of any page, append `.md` to the page URL
- For section-specific indexes, append `/llms.txt` to any section URL
- For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/fastgen/_mcp/server

## Docs

- [NVIDIA FastGen Overview](https://docs.nvidia.com/fastgen.md): Train, distill, and evaluate diffusion models across modalities with PyTorch.
- [Get started](https://docs.nvidia.com/fastgen/get-started.md)
- [Troubleshooting](https://docs.nvidia.com/fastgen/troubleshooting.md): Resolve common FastGen installation, configuration, memory, and checkpoint problems.
- [Release notes](https://docs.nvidia.com/fastgen/release-notes.md): Find FastGen releases and understand the version used for documented benchmarks.
- [Resources](https://docs.nvidia.com/fastgen/resources.md)
- [Training and checkpoints](https://docs.nvidia.com/fastgen/training.md): Configure distributed training, choose run directories, and resume FastGen jobs.
- [Inference and evaluation](https://docs.nvidia.com/fastgen/inference.md)
- [ComfyUI](https://docs.nvidia.com/fastgen/comfyui.md)
- [Choose a method](https://docs.nvidia.com/fastgen/methods.md)
- [Trajectory matching](https://docs.nvidia.com/fastgen/trajectory-matching.md)
- [Distribution matching](https://docs.nvidia.com/fastgen/distribution-matching.md)
- [Fine-tuning](https://docs.nvidia.com/fastgen/fine-tuning.md)
- [Networks and pretrained models](https://docs.nvidia.com/fastgen/networks.md)
- [Datasets](https://docs.nvidia.com/fastgen/datasets.md)
- [Configuration](https://docs.nvidia.com/fastgen/configuration.md)
- [Callbacks](https://docs.nvidia.com/fastgen/callbacks.md)
- [KV cache](https://docs.nvidia.com/fastgen/kv-cache.md)
- [Third-party components](https://docs.nvidia.com/fastgen/third-party.md)
- [Licenses](https://docs.nvidia.com/fastgen/licenses.md)