Llama-3.3-70B-Instruct | NVIDIA NeMo AutoModel

Llama-3.3-70B-Instruct has a checked-in NeMo AutoModel recipe for text generation. The Hugging Face configuration declares the LlamaForCausalLM architecture.

Fine-Tune Llama-3.3-70B-Instruct

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

automodel examples/llm_finetune/llama3_3/llama_3_3_70b_instruct_squad.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=4, pp_size=2, cp_size=1
Nodes	4
Features	Checkpointing (enabled=true), Activation checkpointing (activation_checkpointing=true), Packed sequences (packed_sequence_size=1024)
Advanced	bf16

More Recipes for This Model

Model Reference

Model Architecture

Property	Value
Task	Text generation
Architecture	`LlamaForCausalLM`

Available Models

meta-llama/Llama-3.3-70B-Instruct

LLM fine-tuning guide