DeepSeek

DeepSeek is a series of open-weight language models from DeepSeek AI. The first-generation models (V1/V2) use standard transformer decoder and Multi-head Latent Attention architectures.


Task	Text Generation
Architecture	`DeepseekForCausalLM`
Parameters	7B – 67B
HF Org	deepseek-ai

Available Models

DeepSeek-V2: 236B total, 21B activated (MoE)
DeepSeek-V2-Chat: instruction-tuned variant
DeepSeek-LLM 7B/67B: dense models

Architecture

DeepseekForCausalLM — DeepSeek v1/v2 dense models

Example HF Models

Model	HF ID
DeepSeek LLM 7B Chat	`deepseek-ai/deepseek-llm-7b-chat`
DeepSeek LLM 67B Chat	`deepseek-ai/deepseek-llm-67b-chat`

Try with NeMo AutoModel

Install NeMo AutoModel and follow the fine-tuning guide to configure a recipe for this model.

1. Install (full instructions):

$ pip install nemo-automodel

2. Clone the repo to get example recipes you can adapt:

$ git clone https://github.com/NVIDIA-NeMo/Automodel.git
$ cd Automodel

3. Fine-tune by adapting a base LLM recipe — override the model ID on the CLI:

$ automodel --nproc-per-node=8 examples/llm_finetune/llama3_2/llama3_2_1b_squad.yaml \
>   --model.pretrained_model_name_or_path <MODEL_HF_ID>

Replace <MODEL_HF_ID> with the model ID from Example HF Models above.

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:26.06.00

2. The recipes are at /opt/Automodel/examples/ — navigate there:

$ cd /opt/Automodel

3. Fine-tune:

$ automodel --nproc-per-node=8 examples/llm_finetune/llama3_2/llama3_2_1b_squad.yaml \
>   --model.pretrained_model_name_or_path <MODEL_HF_ID>

See the Installation Guide and LLM Fine-Tuning Guide.

Fine-Tuning

See the LLM Fine-Tuning Guide.