Qwen3#

Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support

We provide recipes for pretraining and fine-tuning Qwen3 models for the following sizes: 0.6B, 1.7B, 4B, 8B, 14B, 32B, 30B-A3B, and 235B-A30B using NeMo 2.0 and NeMo-Run. These recipes configure a run.Partial for one of the nemo.collections.llm api functions introduced in NeMo 2.0. The recipes are hosted in qwen3_600m, qwen3_1p7b, qwen3_4b, qwen3_8b, qwen3_14b, qwen3_32b, qwen3_30b_a3b, and qwen3_235b_a22b.

NeMo 2.0 Pretraining Recipes#

Note

The pretraining recipes use the MockDataModule for the data argument. You are expected to replace the MockDataModule with your own custom dataset.

We provide an example below on how to invoke the default recipe and override the data argument:

from nemo.collections import llm

pretrain = llm.qwen3_8b.pretrain_recipe(
    name="qwen3_8b_pretraining",
    dir=f"/path/to/checkpoints",
    num_nodes=2,
    num_gpus_per_node=8,
)

# # To override the data argument
# dataloader = a_function_that_configures_your_custom_dataset(
#     global_batch_size=global_batch_size,
#     micro_batch_size=micro_batch_size,
#     seq_length=pretrain.model.config.seq_length,
# )
# pretrain.data = dataloader

NeMo 2.0 Fine-tuning Recipes#

Note

The fine-tuning recipes use the SquadDataModule for the data argument. You can replace the SquadDataModule with your custom dataset.

Warning

When using import_ckpt in NeMo 2.0, ensure your script includes if __name__ == "__main__":. Without this, Python’s multiprocessing won’t initialize threads properly, causing a “Failure to acquire lock” error.

To import the HF model and convert to NeMo 2.0 format, run the following command (this only needs to be done once):

from nemo.collections import llm

llm.import_ckpt(model=llm.Qwen3Model(llm.Qwen3Config8B()), source='hf://Qwen/Qwen3-8B')

We provide an example below on how to invoke the default recipe and override the data argument:

from nemo.collections import llm

recipe = llm.qwen3_8b.finetune_recipe(
    name="qwen3_8b_finetuning",
    dir=f"/path/to/checkpoints",
    num_nodes=1,
    num_gpus_per_node=8,
    peft_scheme='lora',  # 'lora', 'none'
    packed_sequence=False,
)

# # To override the data argument
# dataloader = a_function_that_configures_your_custom_dataset(
#     gbs=gbs,
#     mbs=mbs,
#     seq_length=recipe.model.config.seq_length,
# )
# recipe.data = dataloader

By default, the fine-tuning recipe will run LoRA finetuning with LoRA applied to all linear layers in the language model. To fine-tune the entire model without LoRA, set peft_scheme='none' in the recipe argument.

To fine-tune with sequence packing for a higher throughput, set packed_sequence=True. Note that you may need to tune the global batch size in order to achieve similar convergence.

Note

The configuration in the recipes is done using the NeMo-Run run.Config and run.Partial configuration objects. Please review the NeMo-Run documentation to learn more about its configuration and execution system.

Once you have your final configuration ready, you can execute it on any of the NeMo-Run supported executors. The simplest is the local executor, which just runs the pretraining locally in a separate process. You can use it as follows:

import nemo_run as run

run.run(pretrain, executor=run.LocalExecutor())

Additionally, you can also run it directly in the same Python process as follows:

run.run(pretrain, direct=True)
On this page