Qwen3-VL-30B-A3B-Instruct | NVIDIA NeMo AutoModel

Qwen3-VL-30B-A3B-Instruct has a checked-in NeMo AutoModel recipe for image-text-to-text. The Hugging Face configuration declares the Qwen3VLMoeForConditionalGeneration architecture.

Fine-Tune Qwen3-VL-30B-A3B-Instruct

Follow the installation instructions, then run the recipe from the repository root:

automodel examples/vlm_finetune/qwen3/qwen3_vl_moe_30b_te_deepep.yaml --nproc-per-node 8

Choose a Workflow

Goal	Start Here
Run the primary recipe for this model	Use the recipe configuration.
Try another checked-in recipe for this model	Use the alternate recipe.

Configuration

Setting	Configuration
Hardware	-
Strategy	FSDP2; tp_size=1, pp_size=1, cp_size=1, ep_size=8
Nodes	1
Features	Checkpointing (enabled=true), Transformer Engine, DeepEP (dispatcher=deepep)
Advanced	attention=sdpa, linear=te, experts=gmm, dispatcher=deepep

More Recipes for This Model

qwen3_vl_moe_30b_neat_packing.yaml

Model Reference

Model Architecture

Property	Value
Task	Image-text-to-text
Architecture	`Qwen3VLMoeForConditionalGeneration`

Available Models

Qwen/Qwen3-VL-30B-A3B-Instruct

Vision-language model fine-tuning guide