Skip to main content

bench

Synopsis​

ramalama bench [options] model [arg ...]

MODEL TRANSPORTS​

TransportsPrefixWeb Site
URL basedhttps://, http://, file://https://web.site/ai.model, file://tmp/ai.model
HuggingFacehuggingface://, hf://, hf.co/huggingface.co
ModelScopemodelscope://, ms://modelscope.cn
Ollamaollama://ollama.com
rlcrrlcr://ramalama.com
OCI Container Registriesoci://opencontainers.org
Examples: quay.io, Docker Hub,Artifactory

RamaLama defaults to the Ollama registry transport. This default can be overridden in the ramalama.conf file or via the RAMALAMA_TRANSPORTS environment. export RAMALAMA_TRANSPORT=huggingface Changes RamaLama to use huggingface transport.

Modify individual model transports by specifying the huggingface://, oci://, ollama://, https://, http://, file:// prefix to the model.

URL support means if a model is on a web site or even on your local system, you can run it directly.

Options​

--authfile=password​

path of the authentication file for OCI registries

--backend=auto | vulkan | rocm | cuda | sycl | openvino​

GPU backend to use for inference (default: auto).

Available backends depend on the detected GPU hardware.

auto (default): Automatically selects the preferred backend based on your GPU:

  • AMD GPUs: vulkan (Linux/macOS) or rocm (Windows)
  • NVIDIA GPUs: cuda
  • Intel GPUs: vulkan (Linux/macOS) or sycl (Windows); openvino available as explicit option
  • No GPU: vulkan (CPU fallback)

Platform-specific behavior:

  • On Linux/macOS, Vulkan provides broad compatibility and is preferred for AMD and Intel GPUs
  • On Windows, vulkan is not supported on WSL2, so vendor-specific backends (rocm, sycl) are preferred

Explicit backend selection:

  • vulkan: Use Vulkan-based inference (compatible with AMD, Intel, and CPU)
  • rocm: Use AMD ROCm backend (AMD GPUs only)
  • cuda: Use NVIDIA CUDA backend (NVIDIA GPUs only)
  • sycl: Use Intel SYCL/oneAPI backend (Intel GPUs only)
  • openvino: Use Intel OpenVINO backend (Intel GPUs only); uses ghcr.io/ggml-org/llama.cpp:full-openvino

Available choices: The allowed values for --backend are dynamically determined based on your detected GPU hardware. For example, on a system with an AMD GPU, only auto, vulkan, and rocm are available.

Configuration: The default can be overridden in the ramalama.conf file or via the RAMALAMA_BACKEND environment variable.

Examples:

# Use auto-detection (default)
ramalama bench granite

# Force Vulkan backend
ramalama bench --backend vulkan granite

# Compare performance between backends
ramalama bench --backend vulkan granite
ramalama bench --backend rocm granite

--device​

Add a host device to the container. Optional permissions parameter can be used to specify device permissions by combining r for read, w for write, and m for mknod(2).

Example: --device=/dev/dri/renderD128:/dev/xvdc:rwm

The device specification is passed directly to the underlying container engine. See documentation of the supported container engine for more information.

Pass '--device=none' explicitly add no device to the container, eg for running a CPU-only performance comparison.

--env=​

Set environment variables inside of the container.

This option allows arbitrary environment variables that are available for the process to be launched inside of the container. If an environment variable is specified without a value, the container engine checks the host environment for a value and set the variable only if it is set on the host.

--format​

Set the output format of the benchmark results. Options include json and table (default: table).

--help, -h​

show this help message and exit

--image=IMAGE​

OCI container image to run with specified AI model. RamaLama defaults to using images based on the accelerator it discovers. For example: quay.io/ramalama/ramalama. See the table below for all default images. The default image tag is based on the minor version of the RamaLama package. Version 0.18.0 of RamaLama pulls an image with a :0.18 tag from the quay.io/ramalama OCI repository. The --image option overrides this default.

The default can be overridden in the ramalama.conf file or via the RAMALAMA_IMAGE environment variable. export RAMALAMA_IMAGE=quay.io/ramalama/aiimage:1.2 tells RamaLama to use the quay.io/ramalama/aiimage:1.2 image.

Accelerated images:

AcceleratorImage
CPU, Applequay.io/ramalama/ramalama
HIP_VISIBLE_DEVICESquay.io/ramalama/rocm
CUDA_VISIBLE_DEVICESquay.io/ramalama/cuda
ASAHI_VISIBLE_DEVICESquay.io/ramalama/asahi
INTEL_VISIBLE_DEVICESquay.io/ramalama/intel-gpu
ASCEND_VISIBLE_DEVICESquay.io/ramalama/cann
MUSA_VISIBLE_DEVICESquay.io/ramalama/musa

--keep-groups​

pass --group-add keep-groups to podman (default: False) If GPU device on host system is accessible to user via group access, this option leaks the groups into the container.

--name, -n​

name of the container to run the Model in

--network=none​

set the network mode for the container

--ngl​

number of gpu layers, 0 means CPU inferencing, 999 means use max layers (default: -1) The default -1, means use whatever is automatically deemed appropriate (0 or 999)

--oci-runtime​

Override the default OCI runtime used to launch the container. Container engines like Podman and Docker, have their own default oci runtime that they use. Using this option RamaLama will override these defaults.

On Nvidia based GPU systems, RamaLama defaults to using the nvidia-container-runtime. Use this option to override this selection.

--privileged​

By default, RamaLama containers are unprivileged (=false) and cannot, for example, modify parts of the operating system. This is because by de‐ fault a container is only allowed limited access to devices. A "privi‐ leged" container is given the same access to devices as the user launch‐ ing the container, with the exception of virtual consoles (/dev/tty\d+) when running in systemd mode (--systemd=always).

A privileged container turns off the security features that isolate the container from the host. Dropped Capabilities, limited devices, read- only mount points, Apparmor/SELinux separation, and Seccomp filters are all disabled. Due to the disabled security features, the privileged field should almost never be set as containers can easily break out of confinement.

Containers running in a user namespace (e.g., rootless containers) can‐ not have more privileges than the user that launched them.

--pull=policy​

  • always: Always pull the image and throw an error if the pull fails.
  • missing: Only pull the image when it does not exist in the local containers storage. Throw an error if no image is found and the pull fails.
  • never: Never pull the image but use the one from the local containers storage. Throw an error when no image is found.
  • newer: Pull if the image on the registry is newer than the one in the local containers storage. An image is considered to be newer when the digests are different. Comparing the time stamps is prone to errors. Pull errors are suppressed if a local image was found.

--seed=​

Specify seed rather than using random seed model interaction

--selinux=true​

Enable SELinux container separation

--threads, -t​

Maximum number of cpu threads to use. The default is to use half the cores available on this system for the number of threads.

--tls-verify=true​

require HTTPS and verify certificates when contacting OCI registries

Description​

Benchmark specified AI Model.

Examples​

ramalama bench granite3-moe

See Also​

ramalama(1)

Jan 2025, Originally compiled by Eric Curtin <ecurtin@redhat.com>