llama-stack-mirror/llama_toolchain/distribution/distribution.py
Ashwin Bharambe e830814399
Introduce Llama stack distributions (#22)
* Add distribution CLI scaffolding

* More progress towards `llama distribution install`

* getting closer to a distro definition, distro install + configure works

* Distribution server now functioning

* read existing configuration, save enums properly

* Remove inference uvicorn server entrypoint and llama inference CLI command

* updated dependency and client model name

* Improved exception handling

* local imports for faster cli

* undo a typo, add a passthrough distribution

* implement full-passthrough in the server

* add safety adapters, configuration handling, server + clients

* cleanup, moving stuff to common, nuke utils

* Add a Path() wrapper at the earliest place

* fixes

* Bring agentic system api to toolchain

Add adapter dependencies and resolve adapters using a topological sort

* refactor to reduce size of `agentic_system`

* move straggler files and fix some important existing bugs

* ApiSurface -> Api

* refactor a method out

* Adapter -> Provider

* Make each inference provider into its own subdirectory

* installation fixes

* Rename Distribution -> DistributionSpec, simplify RemoteProviders

* dict key instead of attr

* update inference config to take model and not model_dir

* Fix passthrough streaming, send headers properly not part of body :facepalm

* update safety to use model sku ids and not model dirs

* Update cli_reference.md

* minor fixes

* add DistributionConfig, fix a bug in model download

* Make install + start scripts do proper configuration automatically

* Update CLI_reference

* Nuke fp8_requirements, fold fbgemm into common requirements

* Update README, add newline between API surface configurations

* Refactor download functionality out of the Command so can be reused

* Add `llama model download` alias for `llama download`

* Show message about checksum file so users can check themselves

* Simpler intro statements

* get ollama working

* Reduce a bunch of dependencies from toolchain

Some improvements to the distribution install script

* Avoid using `conda run` since it buffers everything

* update dependencies and rely on LLAMA_TOOLCHAIN_DIR for dev purposes

* add validation for configuration input

* resort imports

* make optional subclasses default to yes for configuration

* Remove additional_pip_packages; move deps to providers

* for inline make 8b model the default

* Add scripts to MANIFEST

* allow installing from test.pypi.org

* Fix #2 to help with testing packages

* Must install llama-models at that same version first

* fix PIP_ARGS

---------

Co-authored-by: Hardik Shah <hjshah@fb.com>
Co-authored-by: Hardik Shah <hjshah@meta.com>
2024-08-08 13:38:41 -07:00

86 lines
2.7 KiB
Python

# 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 inspect
from typing import Dict, List
from llama_toolchain.agentic_system.api.endpoints import AgenticSystem
from llama_toolchain.agentic_system.providers import available_agentic_system_providers
from llama_toolchain.inference.api.endpoints import Inference
from llama_toolchain.inference.providers import available_inference_providers
from llama_toolchain.safety.api.endpoints import Safety
from llama_toolchain.safety.providers import available_safety_providers
from .datatypes import (
Api,
ApiEndpoint,
DistributionSpec,
InlineProviderSpec,
ProviderSpec,
)
# These are the dependencies needed by the distribution server.
# `llama-toolchain` is automatically installed by the installation script.
SERVER_DEPENDENCIES = [
"fastapi",
"python-dotenv",
"uvicorn",
]
def distribution_dependencies(distribution: DistributionSpec) -> List[str]:
# only consider InlineProviderSpecs when calculating dependencies
return [
dep
for provider_spec in distribution.provider_specs.values()
if isinstance(provider_spec, InlineProviderSpec)
for dep in provider_spec.pip_packages
] + SERVER_DEPENDENCIES
def api_endpoints() -> Dict[Api, List[ApiEndpoint]]:
apis = {}
protocols = {
Api.inference: Inference,
Api.safety: Safety,
Api.agentic_system: AgenticSystem,
}
for api, protocol in protocols.items():
endpoints = []
protocol_methods = inspect.getmembers(protocol, predicate=inspect.isfunction)
for name, method in protocol_methods:
if not hasattr(method, "__webmethod__"):
continue
webmethod = method.__webmethod__
route = webmethod.route
# use `post` for all methods right now until we fix up the `webmethod` openapi
# annotation and write our own openapi generator
endpoints.append(ApiEndpoint(route=route, method="post", name=name))
apis[api] = endpoints
return apis
def api_providers() -> Dict[Api, Dict[str, ProviderSpec]]:
inference_providers_by_id = {
a.provider_id: a for a in available_inference_providers()
}
safety_providers_by_id = {a.provider_id: a for a in available_safety_providers()}
agentic_system_providers_by_id = {
a.provider_id: a for a in available_agentic_system_providers()
}
return {
Api.inference: inference_providers_by_id,
Api.safety: safety_providers_by_id,
Api.agentic_system: agentic_system_providers_by_id,
}