> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/automodel/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/automodel/_mcp/server.

# gpt-oss-120b

> Fine-tune gpt-oss-120b for text generation with NeMo AutoModel using checked-in recipes and documented distributed settings.

[GPT-OSS 120B](https://huggingface.co/openai/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](/get-started/installation), then run the recipe from the repository root:

```bash
automodel examples/llm_finetune/gpt_oss/gpt_oss_120b.yaml --nproc-per-node 8
```

Use the [Slurm launcher guide](/job-launchers/slurm-cluster) for the multi-node run.

## Choose a Workflow

| Goal                                         | Start Here                                                                                                                                |
| -------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------- |
| Run the primary recipe for this model        | Use the [recipe configuration](https://github.com/NVIDIA-NeMo/Automodel/blob/main/examples/llm_finetune/gpt_oss/gpt_oss_120b.yaml).       |
| Try another checked-in recipe for this model | Use the [alternate recipe](https://github.com/NVIDIA-NeMo/Automodel/blob/main/examples/llm_benchmark/gpt_oss/gptoss_120b_te_deepep.yaml). |

### 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`](https://github.com/NVIDIA-NeMo/Automodel/blob/main/examples/llm_benchmark/gpt_oss/gptoss_120b_te_deepep.yaml)
* [`gptoss_120b_te_deepep_gb200.yaml`](https://github.com/NVIDIA-NeMo/Automodel/blob/main/examples/llm_benchmark/gpt_oss/gptoss_120b_te_deepep_gb200.yaml)
* [`gptoss_120b_te_mxfp8_gb200.yaml`](https://github.com/NVIDIA-NeMo/Automodel/blob/main/examples/llm_benchmark/gpt_oss/gptoss_120b_te_mxfp8_gb200.yaml)

## Model Reference

### Model Architecture

| Property     | Value               |
| ------------ | ------------------- |
| Task         | Text generation     |
| Architecture | `GptOssForCausalLM` |

### Available Models

* [`openai/gpt-oss-120b`](https://huggingface.co/openai/gpt-oss-120b)

## Related Resources

* [LLM fine-tuning guide](/recipes-e2e-examples/sft-peft)