Howto ramalama

install

install on fedora:

sudo dnf install python3-ramalama

install via pypi:

pip install ramalama

install script linux/mac:

curl -fsSL https://ramalama.ai/install.sh | bash

install podman:

sudo apt install podman -y

usage

set variables:

RAMALAMA_CONTAINER_ENGINE=docker
CUDA_VISIBLE_DEVICES="0"

run model ibm granite:

ramalama run granite

pull model openai gpt-oss:

ramalama pull gpt-oss:latest

serve model:

ramalama serve gpt-oss

serve model with vulkan backend:

ramalama serve --image=quay.io/ramalama/ramalama:latest gemma3:4b

serve model with intel-gpu backend:

ramalama serve --image=quay.io/ramalama/intel-gpu:latest gemma3:4b

pull model deekseek-r1:

ramalama pull deepseek

serve model as daemon with llama-stack and other options:

ramalama serve --port 8080 --api llama-stack --name deepseek-service -d deepseek

chat webui for ramalama:

podman run -it --rm --name ramalamastack-ui -p 8501:8501 -e LLAMA_STACK_ENDPOINT=http://host.containers.internal:8080 quay.io/redhat-et/streamlit_client:latest

show container runtime command output without executing it:

ramalama --dryrun run deepseek

stop model service:

ramalama stop deepseek-service

convert specified model to an oci formatted ai model:

ramalama convert ollama://tinyllama:latest oci://quay.io/rhatdan/tiny:latest

create yaml from containers deployed:

podman kube generate containerid -f myapp.yaml

remove all containers running:

podman rm -af

deploy containers using created yaml:

podman play myapp.yaml

references

• https://crfm.stanford.edu/2023/03/13/alpaca.html
• https://www.youtube.com/watch?v=53NZFC-ReWs
• https://github.com/containers/ramalama/tree/main/docs
• https://github.com/containers/ramalama