Qwen3.5-397B-A17B | NVIDIA NeMo AutoModel

Qwen3.5-397B-A17B has a checked-in NeMo AutoModel recipe for image-text-to-text. The Hugging Face configuration declares the Qwen3_5MoeForConditionalGeneration architecture.

Fine-Tune Qwen3.5-397B-A17B

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

automodel examples/vlm_finetune/qwen3_5_moe/qwen3_5_moe_medpix.yaml --nproc-per-node 8

Use the Slurm launcher guide for the multi-node run.

Choose a Workflow

Goal	Start Here
Run the primary recipe for this model	Use the recipe configuration.
Prepare your environment	Follow the installation instructions.

Configuration

Setting	Configuration
Hardware	-
Strategy	FSDP2; tp_size=1, pp_size=8, cp_size=1, ep_size=32
Nodes	32
Features	HybridEP (dispatcher=hybridep)
Advanced	attention=sdpa, linear=torch, experts=gmm, dispatcher=hybridep

Model Reference

Model Architecture

Property	Value
Task	Image-text-to-text
Architecture	`Qwen3_5MoeForConditionalGeneration`

Available Models

Qwen/Qwen3.5-397B-A17B

Vision-language model fine-tuning guide