Qwen3#

Qwen3 is Alibaba Cloud’s third-generation dense language model series, featuring improved reasoning, instruction following, and multilingual capabilities over Qwen2.

Task

Text Generation

Architecture

Qwen3ForCausalLM

Parameters

0.6B – 32B

HF Org

Qwen

Available Models#

  • Qwen3: 0.6B, 1.7B, 4B, 8B, 14B, 32B

Architecture#

  • Qwen3ForCausalLM

Example HF Models#

Model

HF ID

Qwen3 0.6B

Qwen/Qwen3-0.6B

Qwen3 8B

Qwen/Qwen3-8B

Qwen3 32B

Qwen/Qwen3-32B

Example Recipes#

Recipe

Description

qwen3_0p6b_hellaswag.yaml

SFT — Qwen3 0.6B on HellaSwag

qwen3_8b_squad_spark.yaml

SFT — Qwen3 8B on SQuAD (Spark)

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

3. Run the recipe from inside the repo:

automodel --nproc-per-node=8 examples/llm_finetune/qwen/qwen3_0p6b_hellaswag.yaml
Run with Docker

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:26.02.00

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_0p6b_hellaswag.yaml

See the Installation Guide and LLM Fine-Tuning Guide.

Fine-Tuning#

See the LLM Fine-Tuning Guide for full SFT and LoRA instructions.

Hugging Face Model Cards#