Command-R#

Cohere Command-R is a series of enterprise-grade language models optimized for retrieval-augmented generation (RAG) and tool use. Command-R7B uses the updated Cohere2ForCausalLM architecture.

Task

Text Generation

Architecture

CohereForCausalLM / Cohere2ForCausalLM

Parameters

7B – 104B

HF Org

CohereForAI

Available Models#

  • c4ai-command-r-v01: 35B

  • c4ai-command-r-plus: 104B

  • c4ai-command-r7b-12-2024: 7B (Cohere2ForCausalLM)

Architectures#

  • CohereForCausalLM — Command-R v01, Plus

  • Cohere2ForCausalLM — Command-R7B

Example HF Models#

Model

HF ID

Command-R v01

CohereForAI/c4ai-command-r-v01

Command-R7B

CohereForAI/c4ai-command-r7b-12-2024

Example Recipes#

Recipe

Description

cohere_command_r_7b_squad.yaml

SFT — Command-R 7B on SQuAD

cohere_command_r_7b_squad_peft.yaml

LoRA — Command-R 7B on SQuAD

Try with NeMo AutoModel#

1. Install (full instructions):

pip install nemo-automodel

2. Clone the repo to get the example recipes:

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

3. Run the recipe from inside the repo:

automodel --nproc-per-node=8 examples/llm_finetune/cohere/cohere_command_r_7b_squad.yaml
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:26.02.00

2. Navigate to the AutoModel directory (where the recipes are):

cd /opt/Automodel

3. Run the recipe:

automodel --nproc-per-node=8 examples/llm_finetune/cohere/cohere_command_r_7b_squad.yaml

See the Installation Guide and LLM Fine-Tuning Guide.

Fine-Tuning#

See the LLM Fine-Tuning Guide.

Hugging Face Model Cards#