Model CoverageLarge Language Models

Qwen3-Next

View as Markdown

Qwen3-Next is an advanced MoE language model from Alibaba Cloud’s Qwen team designed for high-throughput inference with large total parameter counts and efficient per-token activation.

TaskText Generation (MoE)
ArchitectureQwen3NextForCausalLM
Parameters80B total / 3B active
HF OrgQwen

Available Models

  • Qwen3-Next-80B-A3B: 80B total parameters, 3B activated per token

Architecture

  • Qwen3NextForCausalLM

Example HF Models

ModelHF ID
Qwen3-Next 80B A3B InstructQwen/Qwen3-Next-80B-A3B-Instruct

Example Recipes

RecipeDescription
qwen3_next_te_deepep.yamlSFT — Qwen3-Next with TE + DeepEP

Try with NeMo AutoModel

1. Install (full instructions):

$pip install nemo-automodel

2. Clone the repo to get the example recipes:

$git clone https://github.com/NVIDIA-NeMo/Automodel.git
$cd Automodel

This recipe was validated on 4 nodes × 8 GPUs (32 H100s). See the Launcher Guide for multi-node setup.

3. Run the recipe from inside the repo:

$automodel --nproc-per-node=8 examples/llm_finetune/qwen/qwen3_next_te_deepep.yaml

1. Pull the container and mount a checkpoint directory:

$docker run --gpus all -it --rm \
>  --shm-size=8g \
>  -v $(pwd)/checkpoints:/opt/Automodel/checkpoints \
>  nvcr.io/nvidia/nemo-automodel:26.06.00

2. Navigate to the AutoModel directory (where the recipes are):

$cd /opt/Automodel

3. Run the recipe:

$automodel --nproc-per-node=8 examples/llm_finetune/qwen/qwen3_next_te_deepep.yaml

See the Installation Guide and LLM Fine-Tuning Guide.

Fine-Tuning

See the Large MoE Fine-Tuning Guide.

Hugging Face Model Cards