phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-06 04:34:57 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge 26a490b7fc into a7f9ce9a3a

Browse source
This commit is contained in:
Charlie Doern 2025-09-23 08:55:26 +02:00 • committed by GitHub
commit fa6c5a0ea6
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
95 changed files with 1854 additions and 465 deletions
Download patch file Download diff file Expand all files Collapse all files

200
llama_stack/cli/stack/_show.py Normal file
View file

@ -0,0 +1,200 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import argparse
import os
import sys
import textwrap
from pathlib import Path
import yaml
from prompt_toolkit import prompt
from prompt_toolkit.completion import WordCompleter
from prompt_toolkit.validation import Validator
from termcolor import cprint
from llama_stack.cli.stack.utils import ImageType, available_templates_specs
from llama_stack.core.build import get_provider_dependencies
from llama_stack.core.datatypes import (
BuildConfig,
BuildProvider,
DistributionSpec,
)
from llama_stack.core.distribution import get_provider_registry
from llama_stack.core.stack import replace_env_vars
from llama_stack.log import get_logger
from llama_stack.providers.datatypes import Api
TEMPLATES_PATH = Path(__file__).parent.parent.parent / "templates"
logger = get_logger(name=__name__, category="cli")
# These are the dependencies needed by the distribution server.
# `llama-stack` is automatically installed by the installation script.
SERVER_DEPENDENCIES = [
"aiosqlite",
"fastapi",
"fire",
"httpx",
"uvicorn",
"opentelemetry-sdk",
"opentelemetry-exporter-otlp-proto-http",
]
def run_stack_show_command(args: argparse.Namespace) -> None:
current_venv = os.environ.get("VIRTUAL_ENV")
env_name = args.env_name or current_venv
if args.distro:
available_templates = available_templates_specs()
if args.distro not in available_templates:
cprint(
f"Could not find template {args.distro}. Please run `llama stack show --list-distros` to check out the available templates",
color="red",
file=sys.stderr,
)
sys.exit(1)
build_config = available_templates[args.distro]
# always venv, conda is gone and container is separate.
build_config.image_type = ImageType.VENV.value
elif args.providers:
provider_list: dict[str, list[BuildProvider]] = dict()
for api_provider in args.providers.split(","):
if "=" not in api_provider:
cprint(
"Could not parse `--providers`. Please ensure the list is in the format api1=provider1,api2=provider2",
color="red",
file=sys.stderr,
)
sys.exit(1)
api, provider_type = api_provider.split("=")
providers_for_api = get_provider_registry().get(Api(api), None)
if providers_for_api is None:
cprint(
f"{api} is not a valid API.",
color="red",
file=sys.stderr,
)
sys.exit(1)
if provider_type in providers_for_api:
provider = BuildProvider(
provider_type=provider_type,
module=None,
)
provider_list.setdefault(api, []).append(provider)
else:
cprint(
f"{provider_type} is not a valid provider for the {api} API.",
color="red",
file=sys.stderr,
)
sys.exit(1)
distribution_spec = DistributionSpec(
providers=provider_list,
description=",".join(args.providers),
)
build_config = BuildConfig(image_type=ImageType.VENV.value, distribution_spec=distribution_spec)
elif not args.config and not args.distro:
name = prompt(
"> Enter a name for your Llama Stack (e.g. my-local-stack): ",
validator=Validator.from_callable(
lambda x: len(x) > 0,
error_message="Name cannot be empty, please enter a name",
),
)
image_type = prompt(
"> Enter the image type you want your Llama Stack to be built as (use <TAB> to see options): ",
completer=WordCompleter([e.value for e in ImageType]),
complete_while_typing=True,
validator=Validator.from_callable(
lambda x: x in [e.value for e in ImageType],
error_message="Invalid image type. Use <TAB> to see options",
),
)
env_name = f"llamastack-{name}"
cprint(
textwrap.dedent(
"""
Llama Stack is composed of several APIs working together. Let's select
the provider types (implementations) you want to use for these APIs.
""",
),
color="green",
file=sys.stderr,
)
cprint("Tip: use <TAB> to see options for the providers.\n", color="green", file=sys.stderr)
providers: dict[str, list[BuildProvider]] = dict()
for api, providers_for_api in get_provider_registry().items():
available_providers = [x for x in providers_for_api.keys() if x not in ("remote", "remote::sample")]
if not available_providers:
continue
api_provider = prompt(
f"> Enter provider for API {api.value}: ",
completer=WordCompleter(available_providers),
complete_while_typing=True,
validator=Validator.from_callable(
lambda x: x in available_providers, # noqa: B023 - see https://github.com/astral-sh/ruff/issues/7847
error_message="Invalid provider, use <TAB> to see options",
),
)
string_providers = api_provider.split(" ")
for provider in string_providers:
providers.setdefault(api.value, []).append(BuildProvider(provider_type=provider))
description = prompt(
"\n > (Optional) Enter a short description for your Llama Stack: ",
default="",
)
distribution_spec = DistributionSpec(
providers=providers,
description=description,
)
build_config = BuildConfig(image_type=image_type, distribution_spec=distribution_spec)
else:
with open(args.config) as f:
try:
contents = yaml.safe_load(f)
contents = replace_env_vars(contents)
build_config = BuildConfig(**contents)
build_config.image_type = "venv"
except Exception as e:
cprint(
f"Could not parse config file {args.config}: {e}",
color="red",
file=sys.stderr,
)
sys.exit(1)
print(f"# Dependencies for {args.distro or args.config or env_name}")
normal_deps, special_deps, external_provider_dependencies = get_provider_dependencies(build_config)
normal_deps += SERVER_DEPENDENCIES
# Quote deps with commas
quoted_normal_deps = [quote_if_needed(dep) for dep in normal_deps]
print(f"uv pip install {' '.join(quoted_normal_deps)}")
for special_dep in special_deps:
print(f"uv pip install {quote_if_needed(special_dep)}")
for external_dep in external_provider_dependencies:
print(f"uv pip install {quote_if_needed(external_dep)}")
def quote_if_needed(dep):
# Add quotes if the dependency contains a comma (likely version specifier)
return f"'{dep}'" if "," in dep else dep

100
llama_stack/cli/stack/build.py
View file

@ -1,100 +0,0 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import argparse
import textwrap
from llama_stack.cli.stack.utils import ImageType
from llama_stack.cli.subcommand import Subcommand
class StackBuild(Subcommand):
def __init__(self, subparsers: argparse._SubParsersAction):
super().__init__()
self.parser = subparsers.add_parser(
"build",
prog="llama stack build",
description="Build a Llama stack container",
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
self._add_arguments()
self.parser.set_defaults(func=self._run_stack_build_command)
def _add_arguments(self):
self.parser.add_argument(
"--config",
type=str,
default=None,
help="Path to a config file to use for the build. You can find example configs in llama_stack.cores/**/build.yaml. If this argument is not provided, you will be prompted to enter information interactively",
)
self.parser.add_argument(
"--template",
type=str,
default=None,
help="""(deprecated) Name of the example template config to use for build. You may use `llama stack build --list-distros` to check out the available distributions""",
)
self.parser.add_argument(
"--distro",
"--distribution",
dest="distribution",
type=str,
default=None,
help="""Name of the distribution to use for build. You may use `llama stack build --list-distros` to check out the available distributions""",
)
self.parser.add_argument(
"--list-distros",
"--list-distributions",
action="store_true",
dest="list_distros",
default=False,
help="Show the available distributions for building a Llama Stack distribution",
)
self.parser.add_argument(
"--image-type",
type=str,
help="Image Type to use for the build. If not specified, will use the image type from the template config.",
choices=[e.value for e in ImageType],
default=None, # no default so we can detect if a user specified --image-type and override image_type in the config
)
self.parser.add_argument(
"--image-name",
type=str,
help=textwrap.dedent(
f"""[for image-type={"|".join(e.value for e in ImageType)}] Name of the virtual environment to use for
the build. If not specified, currently active environment will be used if found.
"""
),
default=None,
)
self.parser.add_argument(
"--print-deps-only",
default=False,
action="store_true",
help="Print the dependencies for the stack only, without building the stack",
)
self.parser.add_argument(
"--run",
action="store_true",
default=False,
help="Run the stack after building using the same image type, name, and other applicable arguments",
)
self.parser.add_argument(
"--providers",
type=str,
default=None,
help="Build a config for a list of providers and only those providers. This list is formatted like: api1=provider1,api2=provider2. Where there can be multiple providers per API.",
)
def _run_stack_build_command(self, args: argparse.Namespace) -> None:
# always keep implementation completely silo-ed away from CLI so CLI
# can be fast to load and reduces dependencies
from ._build import run_stack_build_command
return run_stack_build_command(args)

69
llama_stack/cli/stack/show.py Normal file
View file

@ -0,0 +1,69 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import argparse
import textwrap
from llama_stack.cli.stack.utils import ImageType
from llama_stack.cli.subcommand import Subcommand
class StackShow(Subcommand):
def __init__(self, subparsers: argparse._SubParsersAction):
super().__init__()
self.parser = subparsers.add_parser(
"show",
prog="llama stack show",
description="show the dependencies for a llama stack distribution",
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
self._add_arguments()
self.parser.set_defaults(func=self._run_stack_show_command)
def _add_arguments(self):
self.parser.add_argument(
"--config",
type=str,
default=None,
help="Path to a config file to use for the build. You can find example configs in llama_stack/distributions/**/build.yaml. If this argument is not provided, you will be prompted to enter information interactively",
)
self.parser.add_argument(
"--distro",
type=str,
default=None,
help="Name of the distro config to use for show. You may use `llama stack show --list-distros` to check out the available distros",
)
self.parser.add_argument(
"--list-distros",
action="store_true",
default=False,
help="Show the available templates for building a Llama Stack distribution",
)
self.parser.add_argument(
"--env-name",
type=str,
help=textwrap.dedent(
f"""[for image-type={"|".join(e.value for e in ImageType)}] Name of the conda or virtual environment to use for
the build. If not specified, currently active environment will be used if found.
"""
),
default=None,
)
self.parser.add_argument(
"--providers",
type=str,
default=None,
help="sync dependencies for a list of providers and only those providers. This list is formatted like: api1=provider1,api2=provider2. Where there can be multiple providers per API.",
)
def _run_stack_show_command(self, args: argparse.Namespace) -> None:
# always keep implementation completely silo-ed away from CLI so CLI
# can be fast to load and reduces dependencies
from ._show import run_stack_show_command
return run_stack_show_command(args)

4
llama_stack/cli/stack/stack.py
View file

@ -11,11 +11,11 @@ from llama_stack.cli.stack.list_stacks import StackListBuilds
from llama_stack.cli.stack.utils import print_subcommand_description
from llama_stack.cli.subcommand import Subcommand
from .build import StackBuild
from .list_apis import StackListApis
from .list_providers import StackListProviders
from .remove import StackRemove
from .run import StackRun
from .show import StackShow
class StackParser(Subcommand):
@ -39,7 +39,7 @@ class StackParser(Subcommand):
subparsers = self.parser.add_subparsers(title="stack_subcommands")
# Add sub-commands
StackBuild.create(subparsers)
StackShow.create(subparsers)
StackListApis.create(subparsers)
StackListProviders.create(subparsers)
StackRun.create(subparsers)

109
llama_stack/cli/stack/utils.py
View file

@ -4,7 +4,28 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import json
import sys
from enum import Enum
from functools import lru_cache
from pathlib import Path
import yaml
from termcolor import cprint
from llama_stack.core.datatypes import (
BuildConfig,
Provider,
StackRunConfig,
)
from llama_stack.core.distribution import get_provider_registry
from llama_stack.core.resolver import InvalidProviderError
from llama_stack.core.utils.config_dirs import EXTERNAL_PROVIDERS_DIR
from llama_stack.core.utils.dynamic import instantiate_class_type
from llama_stack.core.utils.image_types import LlamaStackImageType
from llama_stack.providers.datatypes import Api
TEMPLATES_PATH = Path(__file__).parent.parent.parent / "distributions"
class ImageType(Enum):
@ -19,3 +40,91 @@ def print_subcommand_description(parser, subparsers):
description = subcommand.description
description_text += f" {name:<21} {description}\n"
parser.epilog = description_text
def generate_run_config(
build_config: BuildConfig,
build_dir: Path,
image_name: str,
) -> Path:
"""
Generate a run.yaml template file for user to edit from a build.yaml file
"""
apis = list(build_config.distribution_spec.providers.keys())
run_config = StackRunConfig(
container_image=(image_name if build_config.image_type == LlamaStackImageType.CONTAINER.value else None),
image_name=image_name,
apis=apis,
providers={},
external_providers_dir=build_config.external_providers_dir
if build_config.external_providers_dir
else EXTERNAL_PROVIDERS_DIR,
)
# build providers dict
provider_registry = get_provider_registry(build_config)
for api in apis:
run_config.providers[api] = []
providers = build_config.distribution_spec.providers[api]
for provider in providers:
pid = provider.provider_type.split("::")[-1]
p = provider_registry[Api(api)][provider.provider_type]
if p.deprecation_error:
raise InvalidProviderError(p.deprecation_error)
try:
config_type = instantiate_class_type(provider_registry[Api(api)][provider.provider_type].config_class)
except (ModuleNotFoundError, ValueError) as exc:
# HACK ALERT:
# This code executes after building is done, the import cannot work since the
# package is either available in the venv or container - not available on the host.
# TODO: use a "is_external" flag in ProviderSpec to check if the provider is
# external
cprint(
f"Failed to import provider {provider.provider_type} for API {api} - assuming it's external, skipping: {exc}",
color="yellow",
file=sys.stderr,
)
# Set config_type to None to avoid UnboundLocalError
config_type = None
if config_type is not None and hasattr(config_type, "sample_run_config"):
config = config_type.sample_run_config(__distro_dir__=f"~/.llama/distributions/{image_name}")
else:
config = {}
p_spec = Provider(
provider_id=pid,
provider_type=provider.provider_type,
config=config,
module=provider.module,
)
run_config.providers[api].append(p_spec)
run_config_file = build_dir / f"{image_name}-run.yaml"
with open(run_config_file, "w") as f:
to_write = json.loads(run_config.model_dump_json())
f.write(yaml.dump(to_write, sort_keys=False))
# Only print this message for non-container builds since it will be displayed before the
# container is built
# For non-container builds, the run.yaml is generated at the very end of the build process so it
# makes sense to display this message
if build_config.image_type != LlamaStackImageType.CONTAINER.value:
cprint(f"You can now run your stack with `llama stack run {run_config_file}`", color="green", file=sys.stderr)
return run_config_file
@lru_cache
def available_templates_specs() -> dict[str, BuildConfig]:
import yaml
template_specs = {}
for p in TEMPLATES_PATH.rglob("*build.yaml"):
template_name = p.parent.name
with open(p) as f:
build_config = BuildConfig(**yaml.safe_load(f))
template_specs[template_name] = build_config
return template_specs

11
llama_stack/core/build.py
View file

@ -6,6 +6,8 @@
import importlib.resources
import sys
import tomllib
from pathlib import Path
from pydantic import BaseModel
from termcolor import cprint
@ -72,8 +74,13 @@ def get_provider_dependencies(
external_provider_deps.append(provider_spec.module)
else:
external_provider_deps.extend(provider_spec.module)
if hasattr(provider_spec, "pip_packages"):
deps.extend(provider_spec.pip_packages)
pyproject = Path(provider_spec.module.replace(".", "/")) / "pyproject.toml"
with open(pyproject, "rb") as f:
data = tomllib.load(f)
dependencies = data.get("project", {}).get("dependencies", [])
deps.extend(dependencies)
if hasattr(provider_spec, "container_image") and provider_spec.container_image:
raise ValueError("A stack's dependencies cannot have a container image")

5
llama_stack/providers/datatypes.py
View file

@ -141,7 +141,10 @@ class ProviderSpec(BaseModel):
)
is_external: bool = Field(default=False, description="Notes whether this provider is an external provider.")
package_extras: list[str] = Field(
default_factory=list,
description="Optional package extras to install when using pyproject.toml files (e.g., ['cpu', 'gpu'])",
)
# used internally by the resolver; this is a hack for now
deps__: list[str] = Field(default_factory=list)

28
llama_stack/providers/inline/agents/meta_reference/pyproject.toml Normal file
View file

@ -0,0 +1,28 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-agents-meta-reference"
version = "0.1.0"
description = "Meta's reference implementation of an agent system that can use tools, access vector databases, and perform complex reasoning tasks"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"matplotlib",
"pillow",
"pandas",
"scikit-learn",
"mcp>=1.8.1",
"aiosqlite",
"psycopg2-binary",
"redis",
"pymongo"
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/inline/batches/reference/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-batches-meta-reference"
version = "0.1.0"
description = "Meta's reference implementation of batches API with KVStore persistence."
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"openai"
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/inline/datasetio/localfs/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-datasetio-localfs"
version = "0.1.0"
description = "Local filesystem-based dataset I/O provider for reading and writing datasets to local storage"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"pandas",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

24
llama_stack/providers/inline/eval/meta_reference/pyproject.toml Normal file
View file

@ -0,0 +1,24 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-eval-meta-reference"
version = "0.1.0"
description = "Meta's reference implementation of evaluation tasks with support for multiple languages and evaluation metrics"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"tree_sitter",
"pythainlp",
"langdetect",
"emoji",
"nltk",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

18
llama_stack/providers/inline/files/localfs/pyproject.toml Normal file
View file

@ -0,0 +1,18 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-files-localfs"
version = "0.1.0"
description = "Local filesystem-based file storage provider for managing files and documents locally"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = []
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

29
llama_stack/providers/inline/inference/meta_reference/pyproject.toml Normal file
View file

@ -0,0 +1,29 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-meta-reference"
version = "0.1.0"
description = "Meta's reference implementation of inference with support for various model formats and optimization techniques"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"accelerate",
"fairscale",
"torch",
"torchvision",
"transformers",
"zmq",
"lm-format-enforcer",
"sentence-transformers",
"torchao==0.8.0",
"fbgemm-gpu-genai==1.1.2",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

22
llama_stack/providers/inline/inference/sentence_transformers/pyproject.toml Normal file
View file

@ -0,0 +1,22 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-sentence-transformers"
version = "0.1.0"
description = "Sentence Transformers inference provider for text embeddings and similarity search"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"torch",
"torchvision",
"sentence-transformers",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

24
llama_stack/providers/inline/post_training/huggingface/pyproject.toml Normal file
View file

@ -0,0 +1,24 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-post-training-huggingface"
version = "0.1.0"
description = "HuggingFace-based post-training provider for fine-tuning models using the HuggingFace ecosystem"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"trl",
"transformers",
"peft",
"datasets>=4.0.0",
"torch",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

30
llama_stack/providers/inline/post_training/torchtune/pyproject.toml Normal file
View file

@ -0,0 +1,30 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-post-training-torchtune"
version = "0.1.0"
description = "TorchTune-based post-training provider for fine-tuning and optimizing models using Meta's TorchTune framework"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"torchtune>=0.5.0",
"torchao>=0.12.0",
"numpy",
]
[project.optional-dependencies]
cpu = [
"torch --extra-index-url https://download.pytorch.org/whl/cpu",
]
gpu = [
"torch",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/inline/safety/code_scanner/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-safety-code-scanner"
version = "0.1.0"
description = "Code Scanner safety provider for detecting security vulnerabilities and unsafe code patterns"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"codeshield",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

18
llama_stack/providers/inline/safety/llama_guard/pyproject.toml Normal file
View file

@ -0,0 +1,18 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-safety-llama-guard"
version = "0.1.0"
description = "Llama Guard safety provider for content moderation and safety filtering using Meta's Llama Guard model"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = []
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/inline/safety/prompt_guard/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-safety-prompt-guard"
version = "0.1.0"
description = "Prompt Guard safety provider for detecting and filtering unsafe prompts and content"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"transformers[accelerate]",
"torch",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/inline/scoring/basic/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-scoring-basic"
version = "0.1.0"
description = "Basic scoring provider for simple evaluation metrics and scoring functions"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"requests",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/inline/scoring/braintrust/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-scoring-braintrust"
version = "0.1.0"
description = "Braintrust scoring provider for evaluation and scoring using the Braintrust platform"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"autoevals",
"openai",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

18
llama_stack/providers/inline/scoring/llm_as_judge/pyproject.toml Normal file
View file

@ -0,0 +1,18 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-scoring-llm-as-judge"
version = "0.1.0"
description = "LLM-as-judge scoring provider that uses language models to evaluate and score responses"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = []
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/inline/telemetry/meta_reference/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-telemetry-meta-reference"
version = "0.1.0"
description = "Meta's reference implementation of telemetry and observability using OpenTelemetry"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"opentelemetry-sdk",
"opentelemetry-exporter-otlp-proto-http",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

28
llama_stack/providers/inline/tool_runtime/rag/pyproject.toml Normal file
View file

@ -0,0 +1,28 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-tool-runtime-rag"
version = "0.1.0"
description = "RAG (Retrieval-Augmented Generation) tool runtime for document ingestion, chunking, and semantic search"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"chardet",
"pypdf",
"tqdm",
"numpy",
"scikit-learn",
"scipy",
"nltk",
"sentencepiece",
"transformers",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/inline/vector_io/chroma/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-vector-io-chroma"
version = "0.1.0"
description = "Chroma inline vector database provider for Llama Stack"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"chromadb",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/inline/vector_io/faiss/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-vector-io-faiss"
version = "0.1.0"
description = "Faiss inline vector database provider for Llama Stack"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"faiss-cpu",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/inline/vector_io/milvus/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-vector-io-milvus"
version = "0.1.0"
description = "Milvus inline vector database provider for Llama Stack"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"pymilvus[milvus-lite]>=2.4.10",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/inline/vector_io/qdrant/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-vector-io-qdrant"
version = "0.1.0"
description = "Qdrant inline vector database provider for Llama Stack"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"qdrant-client",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/inline/vector_io/sqlite_vec/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-vector-io-sqlite-vec"
version = "0.1.0"
description = "SQLite-Vec inline vector database provider for Llama Stack"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"sqlite-vec",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

9
llama_stack/providers/registry/agents.py
View file

@ -10,7 +10,6 @@ from llama_stack.providers.datatypes import (
InlineProviderSpec,
ProviderSpec,
)
from llama_stack.providers.utils.kvstore import kvstore_dependencies
def available_providers() -> list[ProviderSpec]:
@ -18,14 +17,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.agents,
provider_type="inline::meta-reference",
pip_packages=[
"matplotlib",
"pillow",
"pandas",
"scikit-learn",
"mcp>=1.8.1",
]
+ kvstore_dependencies(), # TODO make this dynamic based on the kvstore config
module="llama_stack.providers.inline.agents.meta_reference",
config_class="llama_stack.providers.inline.agents.meta_reference.MetaReferenceAgentsImplConfig",
api_dependencies=[

1
llama_stack/providers/registry/batches.py
View file

@ -13,7 +13,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.batches,
provider_type="inline::reference",
pip_packages=[],
module="llama_stack.providers.inline.batches.reference",
config_class="llama_stack.providers.inline.batches.reference.config.ReferenceBatchesImplConfig",
api_dependencies=[

7
llama_stack/providers/registry/datasetio.py
View file

@ -18,7 +18,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.datasetio,
provider_type="inline::localfs",
pip_packages=["pandas"],
module="llama_stack.providers.inline.datasetio.localfs",
config_class="llama_stack.providers.inline.datasetio.localfs.LocalFSDatasetIOConfig",
api_dependencies=[],
@ -28,9 +27,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.datasetio,
adapter_type="huggingface",
provider_type="remote::huggingface",
pip_packages=[
"datasets>=4.0.0",
],
module="llama_stack.providers.remote.datasetio.huggingface",
config_class="llama_stack.providers.remote.datasetio.huggingface.HuggingfaceDatasetIOConfig",
description="HuggingFace datasets provider for accessing and managing datasets from the HuggingFace Hub.",
@ -41,9 +37,6 @@ def available_providers() -> list[ProviderSpec]:
provider_type="remote::nvidia",
module="llama_stack.providers.remote.datasetio.nvidia",
config_class="llama_stack.providers.remote.datasetio.nvidia.NvidiaDatasetIOConfig",
pip_packages=[
"datasets>=4.0.0",
],
description="NVIDIA's dataset I/O provider for accessing datasets from NVIDIA's data platform.",
),
]

4
llama_stack/providers/registry/eval.py
View file

@ -13,7 +13,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.eval,
provider_type="inline::meta-reference",
pip_packages=["tree_sitter", "pythainlp", "langdetect", "emoji", "nltk"],
module="llama_stack.providers.inline.eval.meta_reference",
config_class="llama_stack.providers.inline.eval.meta_reference.MetaReferenceEvalConfig",
api_dependencies=[
@ -28,9 +27,6 @@ def available_providers() -> list[ProviderSpec]:
RemoteProviderSpec(
api=Api.eval,
adapter_type="nvidia",
pip_packages=[
"requests",
],
provider_type="remote::nvidia",
module="llama_stack.providers.remote.eval.nvidia",
config_class="llama_stack.providers.remote.eval.nvidia.NVIDIAEvalConfig",

1
llama_stack/providers/registry/files.py
View file

@ -14,7 +14,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.files,
provider_type="inline::localfs",
# TODO: make this dynamic according to the sql store type
pip_packages=sql_store_pip_packages,
module="llama_stack.providers.inline.files.localfs",
config_class="llama_stack.providers.inline.files.localfs.config.LocalfsFilesImplConfig",
description="Local filesystem-based file storage provider for managing files and documents locally.",

43
llama_stack/providers/registry/inference.py
View file

@ -31,7 +31,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.inference,
provider_type="inline::meta-reference",
pip_packages=META_REFERENCE_DEPS,
module="llama_stack.providers.inline.inference.meta_reference",
config_class="llama_stack.providers.inline.inference.meta_reference.MetaReferenceInferenceConfig",
description="Meta's reference implementation of inference with support for various model formats and optimization techniques.",
@ -39,11 +38,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.inference,
provider_type="inline::sentence-transformers",
# CrossEncoder depends on torchao.quantization
pip_packages=[
"torch torchvision torchao>=0.12.0 --extra-index-url https://download.pytorch.org/whl/cpu",
"sentence-transformers --no-deps",
],
module="llama_stack.providers.inline.inference.sentence_transformers",
config_class="llama_stack.providers.inline.inference.sentence_transformers.config.SentenceTransformersInferenceConfig",
description="Sentence Transformers inference provider for text embeddings and similarity search.",
@ -52,9 +46,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="cerebras",
provider_type="remote::cerebras",
pip_packages=[
"cerebras_cloud_sdk",
],
module="llama_stack.providers.remote.inference.cerebras",
config_class="llama_stack.providers.remote.inference.cerebras.CerebrasImplConfig",
description="Cerebras inference provider for running models on Cerebras Cloud platform.",
@ -63,7 +54,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="ollama",
provider_type="remote::ollama",
pip_packages=["ollama", "aiohttp", "h11>=0.16.0"],
config_class="llama_stack.providers.remote.inference.ollama.OllamaImplConfig",
module="llama_stack.providers.remote.inference.ollama",
description="Ollama inference provider for running local models through the Ollama runtime.",
@ -72,7 +62,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="vllm",
provider_type="remote::vllm",
pip_packages=[],
module="llama_stack.providers.remote.inference.vllm",
config_class="llama_stack.providers.remote.inference.vllm.VLLMInferenceAdapterConfig",
provider_data_validator="llama_stack.providers.remote.inference.vllm.VLLMProviderDataValidator",
@ -82,7 +71,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="tgi",
provider_type="remote::tgi",
pip_packages=["huggingface_hub", "aiohttp"],
module="llama_stack.providers.remote.inference.tgi",
config_class="llama_stack.providers.remote.inference.tgi.TGIImplConfig",
description="Text Generation Inference (TGI) provider for HuggingFace model serving.",
@ -91,7 +79,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="hf::serverless",
provider_type="remote::hf::serverless",
pip_packages=["huggingface_hub", "aiohttp"],
module="llama_stack.providers.remote.inference.tgi",
config_class="llama_stack.providers.remote.inference.tgi.InferenceAPIImplConfig",
description="HuggingFace Inference API serverless provider for on-demand model inference.",
@ -100,7 +87,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
provider_type="remote::hf::endpoint",
adapter_type="hf::endpoint",
pip_packages=["huggingface_hub", "aiohttp"],
module="llama_stack.providers.remote.inference.tgi",
config_class="llama_stack.providers.remote.inference.tgi.InferenceEndpointImplConfig",
description="HuggingFace Inference Endpoints provider for dedicated model serving.",
@ -109,9 +95,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="fireworks",
provider_type="remote::fireworks",
pip_packages=[
"fireworks-ai<=0.17.16",
],
module="llama_stack.providers.remote.inference.fireworks",
config_class="llama_stack.providers.remote.inference.fireworks.FireworksImplConfig",
provider_data_validator="llama_stack.providers.remote.inference.fireworks.FireworksProviderDataValidator",
@ -121,9 +104,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="together",
provider_type="remote::together",
pip_packages=[
"together",
],
module="llama_stack.providers.remote.inference.together",
config_class="llama_stack.providers.remote.inference.together.TogetherImplConfig",
provider_data_validator="llama_stack.providers.remote.inference.together.TogetherProviderDataValidator",
@ -133,7 +113,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="bedrock",
provider_type="remote::bedrock",
pip_packages=["boto3"],
module="llama_stack.providers.remote.inference.bedrock",
config_class="llama_stack.providers.remote.inference.bedrock.BedrockConfig",
description="AWS Bedrock inference provider for accessing various AI models through AWS's managed service.",
@ -142,7 +121,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="databricks",
provider_type="remote::databricks",
pip_packages=[],
module="llama_stack.providers.remote.inference.databricks",
config_class="llama_stack.providers.remote.inference.databricks.DatabricksImplConfig",
description="Databricks inference provider for running models on Databricks' unified analytics platform.",
@ -151,7 +129,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="nvidia",
provider_type="remote::nvidia",
pip_packages=[],
module="llama_stack.providers.remote.inference.nvidia",
config_class="llama_stack.providers.remote.inference.nvidia.NVIDIAConfig",
description="NVIDIA inference provider for accessing NVIDIA NIM models and AI services.",
@ -160,7 +137,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="runpod",
provider_type="remote::runpod",
pip_packages=[],
module="llama_stack.providers.remote.inference.runpod",
config_class="llama_stack.providers.remote.inference.runpod.RunpodImplConfig",
description="RunPod inference provider for running models on RunPod's cloud GPU platform.",
@ -169,7 +145,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="openai",
provider_type="remote::openai",
pip_packages=["litellm"],
module="llama_stack.providers.remote.inference.openai",
config_class="llama_stack.providers.remote.inference.openai.OpenAIConfig",
provider_data_validator="llama_stack.providers.remote.inference.openai.config.OpenAIProviderDataValidator",
@ -179,7 +154,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="anthropic",
provider_type="remote::anthropic",
pip_packages=["litellm"],
module="llama_stack.providers.remote.inference.anthropic",
config_class="llama_stack.providers.remote.inference.anthropic.AnthropicConfig",
provider_data_validator="llama_stack.providers.remote.inference.anthropic.config.AnthropicProviderDataValidator",
@ -189,9 +163,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="gemini",
provider_type="remote::gemini",
pip_packages=[
"litellm",
],
module="llama_stack.providers.remote.inference.gemini",
config_class="llama_stack.providers.remote.inference.gemini.GeminiConfig",
provider_data_validator="llama_stack.providers.remote.inference.gemini.config.GeminiProviderDataValidator",
@ -201,10 +172,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.inference,
adapter_type="vertexai",
provider_type="remote::vertexai",
pip_packages=[
"litellm",
"google-cloud-aiplatform",
],
module="llama_stack.providers.remote.inference.vertexai",
config_class="llama_stack.providers.remote.inference.vertexai.VertexAIConfig",
provider_data_validator="llama_stack.providers.remote.inference.vertexai.config.VertexAIProviderDataValidator",
@ -233,9 +200,6 @@ Available Models:
api=Api.inference,
adapter_type="groq",
provider_type="remote::groq",
pip_packages=[
"litellm",
],
module="llama_stack.providers.remote.inference.groq",
config_class="llama_stack.providers.remote.inference.groq.GroqConfig",
provider_data_validator="llama_stack.providers.remote.inference.groq.config.GroqProviderDataValidator",
@ -245,7 +209,6 @@ Available Models:
api=Api.inference,
adapter_type="llama-openai-compat",
provider_type="remote::llama-openai-compat",
pip_packages=["litellm"],
module="llama_stack.providers.remote.inference.llama_openai_compat",
config_class="llama_stack.providers.remote.inference.llama_openai_compat.config.LlamaCompatConfig",
provider_data_validator="llama_stack.providers.remote.inference.llama_openai_compat.config.LlamaProviderDataValidator",
@ -255,9 +218,6 @@ Available Models:
api=Api.inference,
adapter_type="sambanova",
provider_type="remote::sambanova",
pip_packages=[
"litellm",
],
module="llama_stack.providers.remote.inference.sambanova",
config_class="llama_stack.providers.remote.inference.sambanova.SambaNovaImplConfig",
provider_data_validator="llama_stack.providers.remote.inference.sambanova.config.SambaNovaProviderDataValidator",
@ -267,7 +227,6 @@ Available Models:
api=Api.inference,
adapter_type="passthrough",
provider_type="remote::passthrough",
pip_packages=[],
module="llama_stack.providers.remote.inference.passthrough",
config_class="llama_stack.providers.remote.inference.passthrough.PassthroughImplConfig",
provider_data_validator="llama_stack.providers.remote.inference.passthrough.PassthroughProviderDataValidator",
@ -277,7 +236,6 @@ Available Models:
api=Api.inference,
adapter_type="watsonx",
provider_type="remote::watsonx",
pip_packages=["ibm_watsonx_ai"],
module="llama_stack.providers.remote.inference.watsonx",
config_class="llama_stack.providers.remote.inference.watsonx.WatsonXConfig",
provider_data_validator="llama_stack.providers.remote.inference.watsonx.WatsonXProviderDataValidator",
@ -287,7 +245,6 @@ Available Models:
api=Api.inference,
provider_type="remote::azure",
adapter_type="azure",
pip_packages=["litellm"],
module="llama_stack.providers.remote.inference.azure",
config_class="llama_stack.providers.remote.inference.azure.AzureConfig",
provider_data_validator="llama_stack.providers.remote.inference.azure.config.AzureProviderDataValidator",

40
llama_stack/providers/registry/post_training.py
View file

@ -5,15 +5,12 @@
# the root directory of this source tree.
from typing import cast
from llama_stack.providers.datatypes import Api, InlineProviderSpec, ProviderSpec, RemoteProviderSpec
# We provide two versions of these providers so that distributions can package the appropriate version of torch.
# The CPU version is used for distributions that don't have GPU support -- they result in smaller container images.
torchtune_def = dict(
api=Api.post_training,
pip_packages=["numpy"],
module="llama_stack.providers.inline.post_training.torchtune",
config_class="llama_stack.providers.inline.post_training.torchtune.TorchtunePostTrainingConfig",
api_dependencies=[
@ -27,28 +24,32 @@ torchtune_def = dict(
def available_providers() -> list[ProviderSpec]:
return [
InlineProviderSpec(
**{ # type: ignore
**torchtune_def,
"provider_type": "inline::torchtune-cpu",
"pip_packages": (
cast(list[str], torchtune_def["pip_packages"])
+ ["torch torchtune>=0.5.0 torchao>=0.12.0 --extra-index-url https://download.pytorch.org/whl/cpu"]
),
},
api=Api.post_training,
provider_type="inline::torchtune-cpu",
module="llama_stack.providers.inline.post_training.torchtune",
config_class="llama_stack.providers.inline.post_training.torchtune.TorchtunePostTrainingConfig",
api_dependencies=[
Api.datasetio,
Api.datasets,
],
description="TorchTune-based post-training provider for fine-tuning and optimizing models using Meta's TorchTune framework (CPU).",
package_extras=["cpu"],
),
InlineProviderSpec(
**{ # type: ignore
**torchtune_def,
"provider_type": "inline::torchtune-gpu",
"pip_packages": (
cast(list[str], torchtune_def["pip_packages"]) + ["torch torchtune>=0.5.0 torchao>=0.12.0"]
),
},
api=Api.post_training,
provider_type="inline::torchtune-gpu",
module="llama_stack.providers.inline.post_training.torchtune",
config_class="llama_stack.providers.inline.post_training.torchtune.TorchtunePostTrainingConfig",
api_dependencies=[
Api.datasetio,
Api.datasets,
],
description="TorchTune-based post-training provider for fine-tuning and optimizing models using Meta's TorchTune framework (GPU).",
package_extras=["gpu"],
),
InlineProviderSpec(
api=Api.post_training,
provider_type="inline::huggingface-gpu",
pip_packages=["trl", "transformers", "peft", "datasets>=4.0.0", "torch"],
module="llama_stack.providers.inline.post_training.huggingface",
config_class="llama_stack.providers.inline.post_training.huggingface.HuggingFacePostTrainingConfig",
api_dependencies=[
@ -61,7 +62,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.post_training,
adapter_type="nvidia",
provider_type="remote::nvidia",
pip_packages=["requests", "aiohttp"],
module="llama_stack.providers.remote.post_training.nvidia",
config_class="llama_stack.providers.remote.post_training.nvidia.NvidiaPostTrainingConfig",
description="NVIDIA's post-training provider for fine-tuning models on NVIDIA's platform.",

11
llama_stack/providers/registry/safety.py
View file

@ -18,10 +18,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.safety,
provider_type="inline::prompt-guard",
pip_packages=[
"transformers[accelerate]",
"torch --index-url https://download.pytorch.org/whl/cpu",
],
module="llama_stack.providers.inline.safety.prompt_guard",
config_class="llama_stack.providers.inline.safety.prompt_guard.PromptGuardConfig",
description="Prompt Guard safety provider for detecting and filtering unsafe prompts and content.",
@ -29,7 +25,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.safety,
provider_type="inline::llama-guard",
pip_packages=[],
module="llama_stack.providers.inline.safety.llama_guard",
config_class="llama_stack.providers.inline.safety.llama_guard.LlamaGuardConfig",
api_dependencies=[
@ -40,9 +35,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.safety,
provider_type="inline::code-scanner",
pip_packages=[
"codeshield",
],
module="llama_stack.providers.inline.safety.code_scanner",
config_class="llama_stack.providers.inline.safety.code_scanner.CodeScannerConfig",
description="Code Scanner safety provider for detecting security vulnerabilities and unsafe code patterns.",
@ -51,7 +43,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.safety,
adapter_type="bedrock",
provider_type="remote::bedrock",
pip_packages=["boto3"],
module="llama_stack.providers.remote.safety.bedrock",
config_class="llama_stack.providers.remote.safety.bedrock.BedrockSafetyConfig",
description="AWS Bedrock safety provider for content moderation using AWS's safety services.",
@ -60,7 +51,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.safety,
adapter_type="nvidia",
provider_type="remote::nvidia",
pip_packages=["requests"],
module="llama_stack.providers.remote.safety.nvidia",
config_class="llama_stack.providers.remote.safety.nvidia.NVIDIASafetyConfig",
description="NVIDIA's safety provider for content moderation and safety filtering.",
@ -69,7 +59,6 @@ def available_providers() -> list[ProviderSpec]:
api=Api.safety,
adapter_type="sambanova",
provider_type="remote::sambanova",
pip_packages=["litellm", "requests"],
module="llama_stack.providers.remote.safety.sambanova",
config_class="llama_stack.providers.remote.safety.sambanova.SambaNovaSafetyConfig",
provider_data_validator="llama_stack.providers.remote.safety.sambanova.config.SambaNovaProviderDataValidator",

3
llama_stack/providers/registry/scoring.py
View file

@ -13,7 +13,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.scoring,
provider_type="inline::basic",
pip_packages=["requests"],
module="llama_stack.providers.inline.scoring.basic",
config_class="llama_stack.providers.inline.scoring.basic.BasicScoringConfig",
api_dependencies=[
@ -25,7 +24,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.scoring,
provider_type="inline::llm-as-judge",
pip_packages=[],
module="llama_stack.providers.inline.scoring.llm_as_judge",
config_class="llama_stack.providers.inline.scoring.llm_as_judge.LlmAsJudgeScoringConfig",
api_dependencies=[
@ -38,7 +36,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.scoring,
provider_type="inline::braintrust",
pip_packages=["autoevals"],
module="llama_stack.providers.inline.scoring.braintrust",
config_class="llama_stack.providers.inline.scoring.braintrust.BraintrustScoringConfig",
api_dependencies=[

4
llama_stack/providers/registry/telemetry.py
View file

@ -17,10 +17,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.telemetry,
provider_type="inline::meta-reference",
pip_packages=[
"opentelemetry-sdk",
"opentelemetry-exporter-otlp-proto-http",
],
optional_api_dependencies=[Api.datasetio],
module="llama_stack.providers.inline.telemetry.meta_reference",
config_class="llama_stack.providers.inline.telemetry.meta_reference.config.TelemetryConfig",

16
llama_stack/providers/registry/tool_runtime.py
View file

@ -18,17 +18,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.tool_runtime,
provider_type="inline::rag-runtime",
pip_packages=[
"chardet",
"pypdf",
"tqdm",
"numpy",
"scikit-learn",
"scipy",
"nltk",
"sentencepiece",
"transformers",
],
module="llama_stack.providers.inline.tool_runtime.rag",
config_class="llama_stack.providers.inline.tool_runtime.rag.config.RagToolRuntimeConfig",
api_dependencies=[Api.vector_io, Api.inference, Api.files],
@ -40,7 +29,6 @@ def available_providers() -> list[ProviderSpec]:
provider_type="remote::brave-search",
module="llama_stack.providers.remote.tool_runtime.brave_search",
config_class="llama_stack.providers.remote.tool_runtime.brave_search.config.BraveSearchToolConfig",
pip_packages=["requests"],
provider_data_validator="llama_stack.providers.remote.tool_runtime.brave_search.BraveSearchToolProviderDataValidator",
description="Brave Search tool for web search capabilities with privacy-focused results.",
),
@ -50,7 +38,6 @@ def available_providers() -> list[ProviderSpec]:
provider_type="remote::bing-search",
module="llama_stack.providers.remote.tool_runtime.bing_search",
config_class="llama_stack.providers.remote.tool_runtime.bing_search.config.BingSearchToolConfig",
pip_packages=["requests"],
provider_data_validator="llama_stack.providers.remote.tool_runtime.bing_search.BingSearchToolProviderDataValidator",
description="Bing Search tool for web search capabilities using Microsoft's search engine.",
),
@ -60,7 +47,6 @@ def available_providers() -> list[ProviderSpec]:
provider_type="remote::tavily-search",
module="llama_stack.providers.remote.tool_runtime.tavily_search",
config_class="llama_stack.providers.remote.tool_runtime.tavily_search.config.TavilySearchToolConfig",
pip_packages=["requests"],
provider_data_validator="llama_stack.providers.remote.tool_runtime.tavily_search.TavilySearchToolProviderDataValidator",
description="Tavily Search tool for AI-optimized web search with structured results.",
),
@ -70,7 +56,6 @@ def available_providers() -> list[ProviderSpec]:
provider_type="remote::wolfram-alpha",
module="llama_stack.providers.remote.tool_runtime.wolfram_alpha",
config_class="llama_stack.providers.remote.tool_runtime.wolfram_alpha.config.WolframAlphaToolConfig",
pip_packages=["requests"],
provider_data_validator="llama_stack.providers.remote.tool_runtime.wolfram_alpha.WolframAlphaToolProviderDataValidator",
description="Wolfram Alpha tool for computational knowledge and mathematical calculations.",
),
@ -80,7 +65,6 @@ def available_providers() -> list[ProviderSpec]:
provider_type="remote::model-context-protocol",
module="llama_stack.providers.remote.tool_runtime.model_context_protocol",
config_class="llama_stack.providers.remote.tool_runtime.model_context_protocol.config.MCPProviderConfig",
pip_packages=["mcp>=1.8.1"],
provider_data_validator="llama_stack.providers.remote.tool_runtime.model_context_protocol.config.MCPProviderDataValidator",
description="Model Context Protocol (MCP) tool for standardized tool calling and context management.",
),

12
llama_stack/providers/registry/vector_io.py
View file

@ -18,7 +18,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.vector_io,
provider_type="inline::meta-reference",
pip_packages=["faiss-cpu"],
module="llama_stack.providers.inline.vector_io.faiss",
config_class="llama_stack.providers.inline.vector_io.faiss.FaissVectorIOConfig",
deprecation_warning="Please use the `inline::faiss` provider instead.",
@ -29,7 +28,6 @@ def available_providers() -> list[ProviderSpec]:
InlineProviderSpec(
api=Api.vector_io,
provider_type="inline::faiss",
pip_packages=["faiss-cpu"],
module="llama_stack.providers.inline.vector_io.faiss",
config_class="llama_stack.providers.inline.vector_io.faiss.FaissVectorIOConfig",
api_dependencies=[Api.inference],
@ -82,7 +80,6 @@ more details about Faiss in general.
InlineProviderSpec(
api=Api.vector_io,
provider_type="inline::sqlite-vec",
pip_packages=["sqlite-vec"],
module="llama_stack.providers.inline.vector_io.sqlite_vec",
config_class="llama_stack.providers.inline.vector_io.sqlite_vec.SQLiteVectorIOConfig",
api_dependencies=[Api.inference],
@ -289,7 +286,6 @@ See [sqlite-vec's GitHub repo](https://github.com/asg017/sqlite-vec/tree/main) f
InlineProviderSpec(
api=Api.vector_io,
provider_type="inline::sqlite_vec",
pip_packages=["sqlite-vec"],
module="llama_stack.providers.inline.vector_io.sqlite_vec",
config_class="llama_stack.providers.inline.vector_io.sqlite_vec.SQLiteVectorIOConfig",
deprecation_warning="Please use the `inline::sqlite-vec` provider (notice the hyphen instead of underscore) instead.",
@ -303,7 +299,6 @@ Please refer to the sqlite-vec provider documentation.
api=Api.vector_io,
adapter_type="chromadb",
provider_type="remote::chromadb",
pip_packages=["chromadb-client"],
module="llama_stack.providers.remote.vector_io.chroma",
config_class="llama_stack.providers.remote.vector_io.chroma.ChromaVectorIOConfig",
api_dependencies=[Api.inference],
@ -345,7 +340,6 @@ See [Chroma's documentation](https://docs.trychroma.com/docs/overview/introducti
InlineProviderSpec(
api=Api.vector_io,
provider_type="inline::chromadb",
pip_packages=["chromadb"],
module="llama_stack.providers.inline.vector_io.chroma",
config_class="llama_stack.providers.inline.vector_io.chroma.ChromaVectorIOConfig",
api_dependencies=[Api.inference],
@ -389,7 +383,6 @@ See [Chroma's documentation](https://docs.trychroma.com/docs/overview/introducti
api=Api.vector_io,
adapter_type="pgvector",
provider_type="remote::pgvector",
pip_packages=["psycopg2-binary"],
module="llama_stack.providers.remote.vector_io.pgvector",
config_class="llama_stack.providers.remote.vector_io.pgvector.PGVectorVectorIOConfig",
api_dependencies=[Api.inference],
@ -500,7 +493,6 @@ See [PGVector's documentation](https://github.com/pgvector/pgvector) for more de
api=Api.vector_io,
adapter_type="weaviate",
provider_type="remote::weaviate",
pip_packages=["weaviate-client"],
module="llama_stack.providers.remote.vector_io.weaviate",
config_class="llama_stack.providers.remote.vector_io.weaviate.WeaviateVectorIOConfig",
provider_data_validator="llama_stack.providers.remote.vector_io.weaviate.WeaviateRequestProviderData",
@ -541,7 +533,6 @@ See [Weaviate's documentation](https://weaviate.io/developers/weaviate) for more
InlineProviderSpec(
api=Api.vector_io,
provider_type="inline::qdrant",
pip_packages=["qdrant-client"],
module="llama_stack.providers.inline.vector_io.qdrant",
config_class="llama_stack.providers.inline.vector_io.qdrant.QdrantVectorIOConfig",
api_dependencies=[Api.inference],
@ -594,7 +585,6 @@ See the [Qdrant documentation](https://qdrant.tech/documentation/) for more deta
api=Api.vector_io,
adapter_type="qdrant",
provider_type="remote::qdrant",
pip_packages=["qdrant-client"],
module="llama_stack.providers.remote.vector_io.qdrant",
config_class="llama_stack.providers.remote.vector_io.qdrant.QdrantVectorIOConfig",
api_dependencies=[Api.inference],
@ -607,7 +597,6 @@ Please refer to the inline provider documentation.
api=Api.vector_io,
adapter_type="milvus",
provider_type="remote::milvus",
pip_packages=["pymilvus>=2.4.10"],
module="llama_stack.providers.remote.vector_io.milvus",
config_class="llama_stack.providers.remote.vector_io.milvus.MilvusVectorIOConfig",
api_dependencies=[Api.inference],
@ -813,7 +802,6 @@ For more details on TLS configuration, refer to the [TLS setup guide](https://mi
InlineProviderSpec(
api=Api.vector_io,
provider_type="inline::milvus",
pip_packages=["pymilvus[milvus-lite]>=2.4.10"],
module="llama_stack.providers.inline.vector_io.milvus",
config_class="llama_stack.providers.inline.vector_io.milvus.MilvusVectorIOConfig",
api_dependencies=[Api.inference],

20
llama_stack/providers/remote/datasetio/huggingface/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-datasetio-huggingface"
version = "0.1.0"
description = "HuggingFace datasets provider for accessing and managing datasets from the HuggingFace Hub"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"datasets",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/datasetio/nvidia/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-datasetio-nvidia"
version = "0.1.0"
description = "NVIDIA's dataset I/O provider for accessing datasets from NVIDIA's data platform"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"datasets",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/eval/nvidia/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-eval-nvidia"
version = "0.1.0"
description = "NVIDIA's evaluation provider for running evaluation tasks on NVIDIA's platform"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"requests",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

23
llama_stack/providers/remote/files/s3/pyproject.toml Normal file
View file

@ -0,0 +1,23 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-files-s3"
version = "0.1.0"
description = "Local filesystem-based file storage provider for managing files and documents locally"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"boto3",
"sqlalchemy[asyncio]",
"aiosqlite",
"asyncpg",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/anthropic/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-anthropic"
version = "0.1.0"
description = "Anthropic inference provider for accessing Claude models and Anthropic's AI services"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"litellm",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

18
llama_stack/providers/remote/inference/azure/pyproject.toml Normal file
View file

@ -0,0 +1,18 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-azure"
version = "0.1.0"
description = "Azure OpenAI inference provider for accessing GPT models and other Azure services. Provider documentation https://learn.microsoft.com/en-us/azure/ai-foundry/openai/overview"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"litellm",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/bedrock/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-bedrock"
version = "0.1.0"
description = "AWS Bedrock inference provider for accessing various AI models through AWS's managed service"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"boto3",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/cerebras/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-cerebras"
version = "0.1.0"
description = "Cerebras inference provider for running models on Cerebras Cloud platform"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"cerebras_cloud_sdk",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/databricks/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-databricks"
version = "0.1.0"
description = "Databricks inference provider for running models on Databricks' unified analytics platform"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"openai",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/fireworks/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-fireworks"
version = "0.1.0"
description = "Fireworks AI inference provider for Llama models and other AI models on the Fireworks platform"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"fireworks-ai",
"grpcio>=1.67.1,<1.71.0", # Pin grpcio version for compatibility
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/gemini/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-gemini"
version = "0.1.0"
description = "Google Gemini inference provider for accessing Gemini models and Google's AI services"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"litellm",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/groq/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-groq"
version = "0.1.0"
description = "Groq inference provider for ultra-fast inference using Groq's LPU technology"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"litellm",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/llama_openai_compat/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-llama-openai-compat"
version = "0.1.0"
description = "Llama OpenAI-compatible provider for using Llama models with OpenAI API format"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"litellm",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/nvidia/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-nvidia"
version = "0.1.0"
description = "NVIDIA inference provider for accessing NVIDIA NIM models and AI services"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"openai",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

23
llama_stack/providers/remote/inference/ollama/pyproject.toml Normal file
View file

@ -0,0 +1,23 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-ollama"
version = "0.1.0"
description = "Ollama inference provider for running local models through the Ollama runtime"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"ollama",
"aiohttp",
"h11>=0.16.0",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/openai/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-openai"
version = "0.1.0"
description = "OpenAI inference provider for accessing GPT models and other OpenAI services"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"litellm",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/inference/passthrough/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-passthrough"
version = "0.1.0"
description = "Passthrough inference provider for connecting to any external inference service not directly supported"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/runpod/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-runpod"
version = "0.1.0"
description = "RunPod inference provider for running models on RunPod's cloud GPU platform"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"openai",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/sambanova/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-sambanova"
version = "0.1.0"
description = "SambaNova inference provider for running models on SambaNova's dataflow architecture"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"litellm",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

22
llama_stack/providers/remote/inference/tgi/pyproject.toml Normal file
View file

@ -0,0 +1,22 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-tgi"
version = "0.1.0"
description = "Text Generation Inference (TGI) provider for HuggingFace model serving"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"huggingface_hub",
"aiohttp",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/together/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-together"
version = "0.1.0"
description = "Together AI inference provider for open-source models and collaborative AI development"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"together",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

19
llama_stack/providers/remote/inference/vertexai/pyproject.toml Normal file
View file

@ -0,0 +1,19 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-vertexai"
version = "0.1.0"
description = "Google VertexAI Remote Inference Provider"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"litellm",
"google-cloud-aiplatform"
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/vllm/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-vllm"
version = "0.1.0"
description = "Remote vLLM inference provider for connecting to vLLM servers"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"openai",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/inference/watsonx/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-inference-watsonx"
version = "0.1.0"
description = "IBM WatsonX inference provider for accessing AI models on IBM's WatsonX platform"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"ibm_watson_machine_learning",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/post_training/nvidia/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-post-training-nvidia"
version = "0.1.0"
description = "NVIDIA's post-training provider for fine-tuning models on NVIDIA's platform"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"requests",
"aiohttp",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/safety/bedrock/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-safety-bedrock"
version = "0.1.0"
description = "AWS Bedrock safety provider for content moderation using AWS's safety services"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"boto3",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/safety/nvidia/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-safety-nvidia"
version = "0.1.0"
description = "NVIDIA's safety provider for content moderation and safety filtering"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"requests",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

21
llama_stack/providers/remote/safety/sambanova/pyproject.toml Normal file
View file

@ -0,0 +1,21 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-safety-sambanova"
version = "0.1.0"
description = "SambaNova's safety provider for content moderation and safety filtering"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"litellm",
"requests",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/tool_runtime/bing_search/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-tool-runtime-bing-search"
version = "0.1.0"
description = "Bing Search tool for web search capabilities using Microsoft's search engine"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"requests",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/tool_runtime/brave_search/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-tool-runtime-brave-search"
version = "0.1.0"
description = "Brave Search tool for web search capabilities with privacy-focused results"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"requests",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/tool_runtime/model_context_protocol/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-tool-runtime-model-context-protocol"
version = "0.1.0"
description = "Model Context Protocol (MCP) tool for standardized tool calling and context management"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"mcp>=1.8.1",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/tool_runtime/tavily_search/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-tool-runtime-tavily-search"
version = "0.1.0"
description = "Tavily Search tool for AI-optimized web search with structured results"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"requests",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/tool_runtime/wolfram_alpha/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-tool-runtime-wolfram-alpha"
version = "0.1.0"
description = "Wolfram Alpha tool for computational knowledge and mathematical calculations"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"requests",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/vector_io/chroma/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-vector-io-chroma-remote"
version = "0.1.0"
description = "Chroma remote vector database provider for Llama Stack"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"chromadb-client",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/vector_io/milvus/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-vector-io-milvus-remote"
version = "0.1.0"
description = "Milvus remote vector database provider for Llama Stack"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"pymilvus>=2.4.10",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/vector_io/pgvector/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-vector-io-pgvector"
version = "0.1.0"
description = "PGVector remote vector database provider for Llama Stack"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"psycopg2-binary",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/vector_io/qdrant/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-vector-io-qdrant-remote"
version = "0.1.0"
description = "Qdrant remote vector database provider for Llama Stack"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"qdrant-client",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]

20
llama_stack/providers/remote/vector_io/weaviate/pyproject.toml Normal file
View file

@ -0,0 +1,20 @@
[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"
[project]
name = "llama-stack-provider-vector-io-weaviate"
version = "0.1.0"
description = "Weaviate remote vector database provider for Llama Stack"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
requires-python = ">=3.12"
license = { "text" = "MIT" }
dependencies = [
"weaviate-client",
]
[tool.setuptools.packages.find]
where = ["."]
include = ["llama_stack*"]