> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/automodel/latest/model-coverage/dllm/z-lab/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/automodel/_mcp/server.

# Z Lab

> Browse NeMo AutoModel coverage for Z Lab diffusion language models.

* [`Qwen3-4B-DFlash-b16`](/model-coverage/dllm/z-lab/Qwen3-4B-DFlash-b16)

> Browse NeMo AutoModel coverage for Z Lab diffusion language models.

## Docs

- [Qwen3-4B-DFlash-b16](https://docs.nvidia.com/nemo/automodel/latest/model-coverage/dllm/z-lab/Qwen3-4B-DFlash-b16.md): Train Qwen3-4B-DFlash-b16 as a speculative-decoding draft model with NeMo AutoModel using the checked-in DFlash recipe and frozen Qwen3 target.