gpt-oss-120b | NVIDIA NeMo AutoModel

GPT-OSS 120B has a checked-in NeMo AutoModel recipe for text generation. The Hugging Face configuration declares the GptOssForCausalLM architecture.

Fine-Tune GPT-OSS 120B

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

automodel examples/llm_finetune/gpt_oss/gpt_oss_120b.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.
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=64
Nodes	8
Features	Checkpointing (enabled=true), HybridEP (dispatcher=hybridep)
Advanced	experts=torch_mm, dispatcher=hybridep

More Recipes for This Model

gptoss_120b_te_deepep.yaml
gptoss_120b_te_deepep_gb200.yaml
gptoss_120b_te_mxfp8_gb200.yaml

Model Reference

Model Architecture

Property	Value
Task	Text generation
Architecture	`GptOssForCausalLM`

Available Models

openai/gpt-oss-120b

LLM fine-tuning guide