More generic image type for OCI-compliant container technologies (#802)

It's a more generic term and applicable to alternatives of Docker, such
as Podman or other OCI-compliant technologies.

---------

Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
This commit is contained in:
Yuan Tang 2025-01-17 19:37:42 -05:00 committed by GitHub
parent 9d005154d7
commit 6da3053c0e
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
30 changed files with 102 additions and 100 deletions

View file

@ -1,6 +1,6 @@
version: '2'
image_name: local
docker_image: null
container_image: null
conda_env: local
apis:
- shields

View file

@ -1,6 +1,6 @@
version: '2'
image_name: local
docker_image: null
container_image: null
conda_env: local
apis:
- shields

View file

@ -1,6 +1,6 @@
version: '2'
image_name: local
docker_image: null
container_image: null
conda_env: local
apis:
- shields

View file

@ -481,7 +481,7 @@
"- telemetry\n",
"conda_env: together\n",
"datasets: <span style=\"font-weight: bold\">[]</span>\n",
"docker_image: null\n",
"container_image: null\n",
"eval_tasks: <span style=\"font-weight: bold\">[]</span>\n",
"image_name: together\n",
"memory_banks: <span style=\"font-weight: bold\">[]</span>\n",
@ -600,7 +600,7 @@
"- telemetry\n",
"conda_env: together\n",
"datasets: \u001b[1m[\u001b[0m\u001b[1m]\u001b[0m\n",
"docker_image: null\n",
"container_image: null\n",
"eval_tasks: \u001b[1m[\u001b[0m\u001b[1m]\u001b[0m\n",
"image_name: together\n",
"memory_banks: \u001b[1m[\u001b[0m\u001b[1m]\u001b[0m\n",

View file

@ -369,7 +369,7 @@
"- telemetry\n",
"- tool_runtime\n",
"datasets: <span style=\"font-weight: bold\">[]</span>\n",
"docker_image: null\n",
"container_image: null\n",
"eval_tasks: <span style=\"font-weight: bold\">[]</span>\n",
"image_name: together\n",
"memory_banks: <span style=\"font-weight: bold\">[]</span>\n",
@ -550,7 +550,7 @@
"- telemetry\n",
"- tool_runtime\n",
"datasets: \u001b[1m[\u001b[0m\u001b[1m]\u001b[0m\n",
"docker_image: null\n",
"container_image: null\n",
"eval_tasks: \u001b[1m[\u001b[0m\u001b[1m]\u001b[0m\n",
"image_name: together\n",
"memory_banks: \u001b[1m[\u001b[0m\u001b[1m]\u001b[0m\n",

View file

@ -760,7 +760,7 @@
"- tool_runtime\n",
"conda_env: together\n",
"datasets: <span style=\"font-weight: bold\">[]</span>\n",
"docker_image: null\n",
"container_image: null\n",
"eval_tasks: <span style=\"font-weight: bold\">[]</span>\n",
"image_name: together\n",
"memory_banks: <span style=\"font-weight: bold\">[]</span>\n",
@ -942,7 +942,7 @@
"- tool_runtime\n",
"conda_env: together\n",
"datasets: \u001b[1m[\u001b[0m\u001b[1m]\u001b[0m\n",
"docker_image: null\n",
"container_image: null\n",
"eval_tasks: \u001b[1m[\u001b[0m\u001b[1m]\u001b[0m\n",
"image_name: together\n",
"memory_banks: \u001b[1m[\u001b[0m\u001b[1m]\u001b[0m\n",

View file

@ -17,13 +17,13 @@ pip install -e .
llama stack build -h
```
We will start build our distribution (in the form of a Conda environment, or Docker image). In this step, we will specify:
We will start build our distribution (in the form of a Conda environment, or Container image). In this step, we will specify:
- `name`: the name for our distribution (e.g. `my-stack`)
- `image_type`: our build image type (`conda | docker`)
- `image_type`: our build image type (`conda | container`)
- `distribution_spec`: our distribution specs for specifying API providers
- `description`: a short description of the configurations for the distribution
- `providers`: specifies the underlying implementation for serving each API endpoint
- `image_type`: `conda` | `docker` to specify whether to build the distribution in the form of Docker image or Conda environment.
- `image_type`: `conda` | `container` to specify whether to build the distribution in the form of Container image or Conda environment.
After this step is complete, a file named `<name>-build.yaml` and template file `<name>-run.yaml` will be generated and saved at the output file path specified at the end of the command.
@ -35,7 +35,7 @@ After this step is complete, a file named `<name>-build.yaml` and template file
llama stack build
> Enter a name for your Llama Stack (e.g. my-local-stack): my-stack
> Enter the image type you want your Llama Stack to be built as (docker or conda): conda
> Enter the image type you want your Llama Stack to be built as (container or conda): conda
Llama Stack is composed of several APIs working together. Let's select
the provider types (implementations) you want to use for these APIs.
@ -348,26 +348,26 @@ llama stack build --config llama_stack/templates/ollama/build.yaml
```
:::
:::{tab-item} Building Docker
:::{tab-item} Building Container
> [!TIP]
> Podman is supported as an alternative to Docker. Set `DOCKER_BINARY` to `podman` in your environment to use Podman.
> Podman is supported as an alternative to Docker. Set `CONTAINER_BINARY` to `podman` in your environment to use Podman.
To build a docker image, you may start off from a template and use the `--image-type docker` flag to specify `docker` as the build image type.
To build a container image, you may start off from a template and use the `--image-type container` flag to specify `container` as the build image type.
```
llama stack build --template ollama --image-type docker
llama stack build --template ollama --image-type container
```
```
$ llama stack build --template ollama --image-type docker
$ llama stack build --template ollama --image-type container
...
Dockerfile created successfully in /tmp/tmp.viA3a3Rdsg/DockerfileFROM python:3.10-slim
Containerfile created successfully in /tmp/tmp.viA3a3Rdsg/ContainerfileFROM python:3.10-slim
...
You can now edit ~/meta-llama/llama-stack/tmp/configs/ollama-run.yaml and run `llama stack run ~/meta-llama/llama-stack/tmp/configs/ollama-run.yaml`
```
After this step is successful, you should be able to find the built docker image and test it with `llama stack run <path/to/run.yaml>`.
After this step is successful, you should be able to find the built container image and test it with `llama stack run <path/to/run.yaml>`.
:::
::::

View file

@ -182,8 +182,8 @@ def _generate_run_config(
"""
apis = list(build_config.distribution_spec.providers.keys())
run_config = StackRunConfig(
docker_image=(
image_name if build_config.image_type == ImageType.docker.value else None
container_image=(
image_name if build_config.image_type == ImageType.container.value else None
),
image_name=image_name,
apis=apis,
@ -238,7 +238,7 @@ def _run_stack_build_command_from_build_config(
image_name: Optional[str] = None,
template_name: Optional[str] = None,
) -> None:
if build_config.image_type == ImageType.docker.value:
if build_config.image_type == ImageType.container.value:
if template_name:
image_name = f"distribution-{template_name}"
else:

View file

@ -47,8 +47,8 @@ class StackBuild(Subcommand):
self.parser.add_argument(
"--image-type",
type=str,
help="Image Type to use for the build. This can be either conda or docker. If not specified, will use the image type from the template config.",
choices=["conda", "docker", "venv"],
help="Image Type to use for the build. This can be either conda or container or venv. If not specified, will use the image type from the template config.",
choices=["conda", "container", "venv"],
default="conda",
)

View file

@ -27,7 +27,7 @@ class StackConfigure(Subcommand):
self.parser.add_argument(
"config",
type=str,
help="Path to the build config file (e.g. ~/.llama/builds/<image_type>/<name>-build.yaml). For docker, this could also be the name of the docker image. ",
help="Path to the build config file (e.g. ~/.llama/builds/<image_type>/<name>-build.yaml). For container, this could also be the name of the container image. ",
)
self.parser.add_argument(

View file

@ -92,9 +92,9 @@ class StackRun(Subcommand):
)
if not config_file.exists() and not has_yaml_suffix:
# check if it's a build config saved to docker dir
# check if it's a build config saved to container dir
config_file = Path(
BUILDS_BASE_DIR / ImageType.docker.value / f"{args.config}-run.yaml"
BUILDS_BASE_DIR / ImageType.container.value / f"{args.config}-run.yaml"
)
if not config_file.exists() and not has_yaml_suffix:
@ -115,12 +115,12 @@ class StackRun(Subcommand):
config_dict = yaml.safe_load(config_file.read_text())
config = parse_and_maybe_upgrade_config(config_dict)
if config.docker_image:
if config.container_image:
script = (
importlib.resources.files("llama_stack")
/ "distribution/start_container.sh"
)
run_args = [script, config.docker_image]
run_args = [script, config.container_image]
else:
current_conda_env = os.environ.get("CONDA_DEFAULT_ENV")
image_name = args.image_name or current_conda_env

View file

@ -38,7 +38,7 @@ SERVER_DEPENDENCIES = [
class ImageType(Enum):
docker = "docker"
container = "container"
conda = "conda"
venv = "venv"
@ -77,8 +77,8 @@ def get_provider_dependencies(
provider_spec = providers_for_api[provider_type]
deps.extend(provider_spec.pip_packages)
if provider_spec.docker_image:
raise ValueError("A stack's dependencies cannot have a docker image")
if provider_spec.container_image:
raise ValueError("A stack's dependencies cannot have a container image")
normal_deps = []
special_deps = []
@ -109,23 +109,25 @@ def build_image(
image_name: str,
template_name: Optional[str] = None,
):
docker_image = build_config.distribution_spec.docker_image or "python:3.10-slim"
container_image = (
build_config.distribution_spec.container_image or "python:3.10-slim"
)
normal_deps, special_deps = get_provider_dependencies(
build_config.distribution_spec.providers
)
normal_deps += SERVER_DEPENDENCIES
if build_config.image_type == ImageType.docker.value:
if build_config.image_type == ImageType.container.value:
script = str(
importlib.resources.files("llama_stack") / "distribution/build_container.sh"
)
args = [
script,
image_name,
docker_image,
container_image,
str(build_file_path),
str(BUILDS_BASE_DIR / ImageType.docker.value),
str(BUILDS_BASE_DIR / ImageType.container.value),
" ".join(normal_deps),
]
elif build_config.image_type == ImageType.conda.value:

View file

@ -13,7 +13,7 @@ PYPI_VERSION=${PYPI_VERSION:-}
BUILD_PLATFORM=${BUILD_PLATFORM:-}
if [ "$#" -lt 4 ]; then
echo "Usage: $0 <build_name> <docker_base> <pip_dependencies> [<special_pip_deps>]" >&2
echo "Usage: $0 <build_name> <container_base> <pip_dependencies> [<special_pip_deps>]" >&2
echo "Example: $0 my-fastapi-app python:3.9-slim 'fastapi uvicorn' " >&2
exit 1
fi
@ -24,7 +24,7 @@ set -euo pipefail
build_name="$1"
image_name="distribution-$build_name"
docker_base=$2
container_base=$2
build_file_path=$3
host_build_dir=$4
pip_dependencies=$5
@ -36,14 +36,14 @@ NC='\033[0m' # No Color
SCRIPT_DIR=$(dirname "$(readlink -f "$0")")
REPO_DIR=$(dirname $(dirname "$SCRIPT_DIR"))
DOCKER_BINARY=${DOCKER_BINARY:-docker}
DOCKER_OPTS=${DOCKER_OPTS:-}
CONTAINER_BINARY=${CONTAINER_BINARY:-docker}
CONTAINER_OPTS=${CONTAINER_OPTS:-}
TEMP_DIR=$(mktemp -d)
add_to_docker() {
add_to_container() {
local input
output_file="$TEMP_DIR/Dockerfile"
output_file="$TEMP_DIR/Containerfile"
if [ -t 0 ]; then
printf '%s\n' "$1" >>"$output_file"
else
@ -53,9 +53,9 @@ add_to_docker() {
}
# Update and install UBI9 components if UBI9 base image is used
if [[ $docker_base == *"registry.access.redhat.com/ubi9"* ]]; then
add_to_docker << EOF
FROM $docker_base
if [[ $container_base == *"registry.access.redhat.com/ubi9"* ]]; then
add_to_container << EOF
FROM $container_base
WORKDIR /app
RUN microdnf -y update && microdnf install -y iputils net-tools wget \
@ -64,8 +64,8 @@ RUN microdnf -y update && microdnf install -y iputils net-tools wget \
EOF
else
add_to_docker << EOF
FROM $docker_base
add_to_container << EOF
FROM $container_base
WORKDIR /app
RUN apt-get update && apt-get install -y \
@ -82,7 +82,7 @@ fi
# Add pip dependencies first since llama-stack is what will change most often
# so we can reuse layers.
if [ -n "$pip_dependencies" ]; then
add_to_docker << EOF
add_to_container << EOF
RUN pip install --no-cache $pip_dependencies
EOF
fi
@ -90,7 +90,7 @@ fi
if [ -n "$special_pip_deps" ]; then
IFS='#' read -ra parts <<<"$special_pip_deps"
for part in "${parts[@]}"; do
add_to_docker <<EOF
add_to_container <<EOF
RUN pip install --no-cache $part
EOF
done
@ -108,16 +108,16 @@ if [ -n "$LLAMA_STACK_DIR" ]; then
# Install in editable format. We will mount the source code into the container
# so that changes will be reflected in the container without having to do a
# rebuild. This is just for development convenience.
add_to_docker << EOF
add_to_container << EOF
RUN pip install --no-cache -e $stack_mount
EOF
else
if [ -n "$TEST_PYPI_VERSION" ]; then
# these packages are damaged in test-pypi, so install them first
add_to_docker << EOF
add_to_container << EOF
RUN pip install fastapi libcst
EOF
add_to_docker << EOF
add_to_container << EOF
RUN pip install --no-cache --extra-index-url https://test.pypi.org/simple/ \
llama-models==$TEST_PYPI_VERSION llama-stack-client==$TEST_PYPI_VERSION llama-stack==$TEST_PYPI_VERSION
@ -128,7 +128,7 @@ EOF
else
SPEC_VERSION="llama-stack"
fi
add_to_docker << EOF
add_to_container << EOF
RUN pip install --no-cache $SPEC_VERSION
EOF
fi
@ -140,14 +140,14 @@ if [ -n "$LLAMA_MODELS_DIR" ]; then
exit 1
fi
add_to_docker << EOF
add_to_container << EOF
RUN pip uninstall -y llama-models
RUN pip install --no-cache $models_mount
EOF
fi
add_to_docker << EOF
add_to_container << EOF
# This would be good in production but for debugging flexibility lets not add it right now
# We need a more solid production ready entrypoint.sh anyway
@ -156,8 +156,8 @@ ENTRYPOINT ["python", "-m", "llama_stack.distribution.server.server", "--templat
EOF
printf "Dockerfile created successfully in $TEMP_DIR/Dockerfile\n\n"
cat $TEMP_DIR/Dockerfile
printf "Containerfile created successfully in $TEMP_DIR/Containerfile\n\n"
cat $TEMP_DIR/Containerfile
printf "\n"
mounts=""
@ -170,7 +170,7 @@ fi
if command -v selinuxenabled &>/dev/null && selinuxenabled; then
# Disable SELinux labels -- we don't want to relabel the llama-stack source dir
DOCKER_OPTS="$DOCKER_OPTS --security-opt label=disable"
CONTAINER_OPTS="$CONTAINER_OPTS --security-opt label=disable"
fi
# Set version tag based on PyPI version
@ -200,7 +200,7 @@ else
fi
set -x
$DOCKER_BINARY build $DOCKER_OPTS $PLATFORM -t $image_tag -f "$TEMP_DIR/Dockerfile" "$REPO_DIR" $mounts
$CONTAINER_BINARY build $CONTAINER_OPTS $PLATFORM -t $image_tag -f "$TEMP_DIR/Containerfile" "$REPO_DIR" $mounts
# clean up tmp/configs
set +x

View file

@ -6,8 +6,8 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
DOCKER_BINARY=${DOCKER_BINARY:-docker}
DOCKER_OPTS=${DOCKER_OPTS:-}
CONTAINER_BINARY=${CONTAINER_BINARY:-docker}
CONTAINER_OPTS=${CONTAINER_OPTS:-}
LLAMA_STACK_DIR=${LLAMA_STACK_DIR:-}
set -euo pipefail
@ -24,13 +24,13 @@ if [ $# -lt 2 ]; then
exit 1
fi
docker_image="$1"
container_image="$1"
host_build_dir="$2"
container_build_dir="/app/builds"
if command -v selinuxenabled &> /dev/null && selinuxenabled; then
# Disable SELinux labels
DOCKER_OPTS="$DOCKER_OPTS --security-opt label=disable"
CONTAINER_OPTS="$CONTAINER_OPTS --security-opt label=disable"
fi
mounts=""
@ -39,9 +39,9 @@ if [ -n "$LLAMA_STACK_DIR" ]; then
fi
set -x
$DOCKER_BINARY run $DOCKER_OPTS -it \
$CONTAINER_BINARY run $CONTAINER_OPTS -it \
--entrypoint "/usr/local/bin/llama" \
-v $host_build_dir:$container_build_dir \
$mounts \
$docker_image \
$container_image \
stack configure ./llamastack-build.yaml --output-dir $container_build_dir

View file

@ -73,7 +73,7 @@ class AutoRoutedProviderSpec(ProviderSpec):
provider_type: str = "router"
config_class: str = ""
docker_image: Optional[str] = None
container_image: Optional[str] = None
routing_table_api: Api
module: str
provider_data_validator: Optional[str] = Field(
@ -89,7 +89,7 @@ class AutoRoutedProviderSpec(ProviderSpec):
class RoutingTableProviderSpec(ProviderSpec):
provider_type: str = "routing_table"
config_class: str = ""
docker_image: Optional[str] = None
container_image: Optional[str] = None
router_api: Api
module: str
@ -101,7 +101,7 @@ class DistributionSpec(BaseModel):
default="",
description="Description of the distribution",
)
docker_image: Optional[str] = None
container_image: Optional[str] = None
providers: Dict[str, Union[str, List[str]]] = Field(
default_factory=dict,
description="""
@ -127,9 +127,9 @@ Reference to the distribution this package refers to. For unregistered (adhoc) p
this could be just a hash
""",
)
docker_image: Optional[str] = Field(
container_image: Optional[str] = Field(
default=None,
description="Reference to the docker image if this package refers to a container",
description="Reference to the container image if this package refers to a container",
)
apis: List[str] = Field(
default_factory=list,
@ -168,5 +168,5 @@ class BuildConfig(BaseModel):
)
image_type: str = Field(
default="conda",
description="Type of package to build (conda | docker | venv)",
description="Type of package to build (conda | container | venv)",
)

View file

@ -6,8 +6,8 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
DOCKER_BINARY=${DOCKER_BINARY:-docker}
DOCKER_OPTS=${DOCKER_OPTS:-}
CONTAINER_BINARY=${CONTAINER_BINARY:-docker}
CONTAINER_OPTS=${CONTAINER_OPTS:-}
LLAMA_CHECKPOINT_DIR=${LLAMA_CHECKPOINT_DIR:-}
LLAMA_STACK_DIR=${LLAMA_STACK_DIR:-}
TEST_PYPI_VERSION=${TEST_PYPI_VERSION:-}
@ -31,7 +31,7 @@ if [ $# -lt 3 ]; then
fi
build_name="$1"
docker_image="localhost/distribution-$build_name"
container_image="localhost/distribution-$build_name"
shift
yaml_config="$1"
@ -64,7 +64,7 @@ set -x
if command -v selinuxenabled &> /dev/null && selinuxenabled; then
# Disable SELinux labels
DOCKER_OPTS="$DOCKER_OPTS --security-opt label=disable"
CONTAINER_OPTS="$CONTAINER_OPTS --security-opt label=disable"
fi
mounts=""
@ -73,7 +73,7 @@ if [ -n "$LLAMA_STACK_DIR" ]; then
fi
if [ -n "$LLAMA_CHECKPOINT_DIR" ]; then
mounts="$mounts -v $LLAMA_CHECKPOINT_DIR:/root/.llama"
DOCKER_OPTS="$DOCKER_OPTS --gpus=all"
CONTAINER_OPTS="$CONTAINER_OPTS --gpus=all"
fi
version_tag="latest"
@ -85,11 +85,11 @@ elif [ -n "$TEST_PYPI_VERSION" ]; then
version_tag="test-$TEST_PYPI_VERSION"
fi
$DOCKER_BINARY run $DOCKER_OPTS -it \
$CONTAINER_BINARY run $CONTAINER_OPTS -it \
-p $port:$port \
$env_vars \
-v "$yaml_config:/app/config.yaml" \
$mounts \
--env LLAMA_STACK_PORT=$port \
--entrypoint='["python", "-m", "llama_stack.distribution.server.server", "--yaml-config", "/app/config.yaml"]' \
$docker_image:$version_tag
$container_image:$version_tag

View file

@ -150,11 +150,11 @@ class InlineProviderSpec(ProviderSpec):
default_factory=list,
description="The pip dependencies needed for this implementation",
)
docker_image: Optional[str] = Field(
container_image: Optional[str] = Field(
default=None,
description="""
The docker image to use for this implementation. If one is provided, pip_packages will be ignored.
If a provider depends on other providers, the dependencies MUST NOT specify a docker image.
The container image to use for this implementation. If one is provided, pip_packages will be ignored.
If a provider depends on other providers, the dependencies MUST NOT specify a container image.
""",
)
module: str = Field(
@ -197,7 +197,7 @@ API responses, specify the adapter here.
)
@property
def docker_image(self) -> Optional[str]:
def container_image(self) -> Optional[str]:
return None
@property

View file

@ -70,7 +70,7 @@ def get_distribution_template() -> DistributionTemplate:
name=name,
distro_type="self_hosted",
description="Use AWS Bedrock for running LLM inference and safety",
docker_image=None,
container_image=None,
template_path=Path(__file__).parent / "doc_template.md",
providers=providers,
default_models=default_models,

View file

@ -92,7 +92,7 @@ def get_distribution_template() -> DistributionTemplate:
name="cerebras",
distro_type="self_hosted",
description="Use Cerebras for running LLM inference",
docker_image=None,
container_image=None,
template_path=Path(__file__).parent / "doc_template.md",
providers=providers,
default_models=default_models,

View file

@ -2,7 +2,7 @@ version: '2'
name: experimental-post-training
distribution_spec:
description: Experimental template for post training
docker_image: null
container_image: null
providers:
inference:
- inline::meta-reference

View file

@ -1,6 +1,6 @@
version: '2'
image_name: experimental-post-training
docker_image: null
container_image: null
conda_env: experimental-post-training
apis:
- agents

View file

@ -98,7 +98,7 @@ def get_distribution_template() -> DistributionTemplate:
name=name,
distro_type="self_hosted",
description="Use Fireworks.AI for running LLM inference",
docker_image=None,
container_image=None,
template_path=Path(__file__).parent / "doc_template.md",
providers=providers,
default_models=default_models,

View file

@ -88,7 +88,7 @@ def get_distribution_template() -> DistributionTemplate:
name=name,
distro_type="self_hosted",
description="Use (an external) Hugging Face Inference Endpoint for running LLM inference",
docker_image=None,
container_image=None,
template_path=None,
providers=providers,
default_models=[inference_model, safety_model],

View file

@ -89,7 +89,7 @@ def get_distribution_template() -> DistributionTemplate:
name=name,
distro_type="self_hosted",
description="Use (an external) Hugging Face Inference Endpoint for running LLM inference",
docker_image=None,
container_image=None,
template_path=None,
providers=providers,
default_models=[inference_model, safety_model],

View file

@ -68,7 +68,7 @@ def get_distribution_template() -> DistributionTemplate:
name="nvidia",
distro_type="remote_hosted",
description="Use NVIDIA NIM for running LLM inference",
docker_image=None,
container_image=None,
template_path=Path(__file__).parent / "doc_template.md",
providers=providers,
default_models=default_models,

View file

@ -90,7 +90,7 @@ def get_distribution_template() -> DistributionTemplate:
name=name,
distro_type="self_hosted",
description="Use (an external) Ollama server for running LLM inference",
docker_image=None,
container_image=None,
template_path=Path(__file__).parent / "doc_template.md",
providers=providers,
default_models=[inference_model, safety_model],

View file

@ -37,7 +37,7 @@ class RunConfigSettings(BaseModel):
self,
name: str,
providers: Dict[str, List[str]],
docker_image: Optional[str] = None,
container_image: Optional[str] = None,
) -> StackRunConfig:
provider_registry = get_provider_registry()
@ -83,7 +83,7 @@ class RunConfigSettings(BaseModel):
return StackRunConfig(
image_name=name,
docker_image=docker_image,
container_image=container_image,
conda_env=name,
apis=apis,
providers=provider_configs,
@ -113,7 +113,7 @@ class DistributionTemplate(BaseModel):
# Optional configuration
run_config_env_vars: Optional[Dict[str, Tuple[str, str]]] = None
docker_image: Optional[str] = None
container_image: Optional[str] = None
default_models: Optional[List[ModelInput]] = None
@ -122,7 +122,7 @@ class DistributionTemplate(BaseModel):
name=self.name,
distribution_spec=DistributionSpec(
description=self.description,
docker_image=self.docker_image,
container_image=self.container_image,
providers=self.providers,
),
image_type="conda", # default to conda, can be overridden
@ -170,7 +170,7 @@ class DistributionTemplate(BaseModel):
for yaml_pth, settings in self.run_configs.items():
run_config = settings.run_config(
self.name, self.providers, self.docker_image
self.name, self.providers, self.container_image
)
with open(yaml_output_dir / yaml_pth, "w") as f:
yaml.safe_dump(

View file

@ -92,7 +92,7 @@ def get_distribution_template() -> DistributionTemplate:
name=name,
distro_type="self_hosted",
description="Use (an external) TGI server for running LLM inference",
docker_image=None,
container_image=None,
template_path=Path(__file__).parent / "doc_template.md",
providers=providers,
default_models=[inference_model, safety_model],

View file

@ -96,7 +96,7 @@ def get_distribution_template() -> DistributionTemplate:
name=name,
distro_type="self_hosted",
description="Use Together.AI for running LLM inference",
docker_image=None,
container_image=None,
template_path=Path(__file__).parent / "doc_template.md",
providers=providers,
default_models=default_models,

View file

@ -84,7 +84,7 @@ def get_distribution_template() -> DistributionTemplate:
name=name,
distro_type="self_hosted",
description="Use a built-in vLLM engine for running LLM inference",
docker_image=None,
container_image=None,
template_path=None,
providers=providers,
default_models=[inference_model],