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

# Qwen3.8-27B

> Fine-tune Qwen3.8-27B for vision-language tasks with NeMo AutoModel using checked-in full-parameter and LoRA recipes with distributed settings.

Qwen3.8-27B is Qwen's dense 27B native vision-language model for text, image,
and video inputs. It retains the Qwen3.5 hybrid Gated DeltaNet and gated-attention
architecture, so NeMo AutoModel uses its existing native
`Qwen3_5ForConditionalGeneration` implementation for the checkpoint.

## Fine-Tune Qwen3.8-27B

```bash
uv run automodel --nproc-per-node=8 examples/vlm_finetune/qwen3_8/qwen3_8_27b.yaml
uv run automodel --nproc-per-node=8 examples/vlm_finetune/qwen3_8/qwen3_8_27b_lora.yaml
```

Install the `vlm-media` extra before running either recipe. See the
[Installation Guide](/get-started/installation) and the
[Qwen3.8 fine-tuning guide](/recipes-e2e-examples/qwen3-8-27b) for setup,
validation details, and measured training-loss curves.

The checked-in recipes use FSDP2 data parallelism on eight GPUs. The shared
Qwen3.5 dense model implementation also supports TP, CP, and PP, but the
Qwen3.8 runs documented here validate the DP8 recipes specifically.

## Model Reference

### Model Architecture

| Property           | Value                               |
| ------------------ | ----------------------------------- |
| **Task**           | Image-Text-to-Text                  |
| **Architecture**   | `Qwen3_5ForConditionalGeneration`   |
| **Parameters**     | 27B dense                           |
| **Context length** | 262,144 tokens                      |
| **HF Org**         | [Qwen](https://huggingface.co/Qwen) |

## Example Models and Recipes

| Model                                                         | Recipe                                                                                                                             | Description                      |
| ------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------- | -------------------------------- |
| [`Qwen/Qwen3.8-27B`](https://huggingface.co/Qwen/Qwen3.8-27B) | [qwen3\_8\_27b.yaml](https://github.com/NVIDIA-NeMo/Automodel/blob/main/examples/vlm_finetune/qwen3_8/qwen3_8_27b.yaml)            | Full-parameter SFT on MedPix-VQA |
| [`Qwen/Qwen3.8-27B`](https://huggingface.co/Qwen/Qwen3.8-27B) | [qwen3\_8\_27b\_lora.yaml](https://github.com/NVIDIA-NeMo/Automodel/blob/main/examples/vlm_finetune/qwen3_8/qwen3_8_27b_lora.yaml) | LoRA on MedPix-VQA               |