mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-07-08 06:44:32 +00:00
[API Updates] Model / shield / memory-bank routing + agent persistence + support for private headers (#92)
This is yet another of those large PRs (hopefully we will have less and less of them as things mature fast). This one introduces substantial improvements and some simplifications to the stack. Most important bits: * Agents reference implementation now has support for session / turn persistence. The default implementation uses sqlite but there's also support for using Redis. * We have re-architected the structure of the Stack APIs to allow for more flexible routing. The motivating use cases are: - routing model A to ollama and model B to a remote provider like Together - routing shield A to local impl while shield B to a remote provider like Bedrock - routing a vector memory bank to Weaviate while routing a keyvalue memory bank to Redis * Support for provider specific parameters to be passed from the clients. A client can pass data using `x_llamastack_provider_data` parameter which can be type-checked and provided to the Adapter implementations.
This commit is contained in:
Ashwin Bharambe
GitHub
parent
8bf8c07eb3
commit
ec4fc800cc
130 changed files with 9701 additions and 11227 deletions
|
|
@ -1,5 +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.
|
||||
|
|
@ -1,15 +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.
|
||||
|
||||
from .config import RedisImplConfig
|
||||
|
||||
|
||||
async def get_adapter_impl(config: RedisImplConfig, _deps):
|
||||
from .redis import RedisControlPlaneAdapter
|
||||
|
||||
impl = RedisControlPlaneAdapter(config)
|
||||
await impl.initialize()
|
||||
return impl
|
||||
|
|
@ -1,21 +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.
|
||||
|
||||
from typing import Optional
|
||||
|
||||
from llama_models.schema_utils import json_schema_type
|
||||
from pydantic import BaseModel, Field
|
||||
|
||||
|
||||
@json_schema_type
|
||||
class RedisImplConfig(BaseModel):
|
||||
url: str = Field(
|
||||
description="The URL for the Redis server",
|
||||
)
|
||||
namespace: Optional[str] = Field(
|
||||
default=None,
|
||||
description="All keys will be prefixed with this namespace",
|
||||
)
|
||||
|
|
@ -1,62 +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.
|
||||
|
||||
from datetime import datetime, timedelta
|
||||
from typing import Any, List, Optional
|
||||
|
||||
from redis.asyncio import Redis
|
||||
|
||||
from llama_stack.apis.control_plane import * # noqa: F403
|
||||
|
||||
|
||||
from .config import RedisImplConfig
|
||||
|
||||
|
||||
class RedisControlPlaneAdapter(ControlPlane):
|
||||
def __init__(self, config: RedisImplConfig):
|
||||
self.config = config
|
||||
|
||||
async def initialize(self) -> None:
|
||||
self.redis = Redis.from_url(self.config.url)
|
||||
|
||||
def _namespaced_key(self, key: str) -> str:
|
||||
if not self.config.namespace:
|
||||
return key
|
||||
return f"{self.config.namespace}:{key}"
|
||||
|
||||
async def set(
|
||||
self, key: str, value: Any, expiration: Optional[datetime] = None
|
||||
) -> None:
|
||||
key = self._namespaced_key(key)
|
||||
await self.redis.set(key, value)
|
||||
if expiration:
|
||||
await self.redis.expireat(key, expiration)
|
||||
|
||||
async def get(self, key: str) -> Optional[ControlPlaneValue]:
|
||||
key = self._namespaced_key(key)
|
||||
value = await self.redis.get(key)
|
||||
if value is None:
|
||||
return None
|
||||
ttl = await self.redis.ttl(key)
|
||||
expiration = datetime.now() + timedelta(seconds=ttl) if ttl > 0 else None
|
||||
return ControlPlaneValue(key=key, value=value, expiration=expiration)
|
||||
|
||||
async def delete(self, key: str) -> None:
|
||||
key = self._namespaced_key(key)
|
||||
await self.redis.delete(key)
|
||||
|
||||
async def range(self, start_key: str, end_key: str) -> List[ControlPlaneValue]:
|
||||
start_key = self._namespaced_key(start_key)
|
||||
end_key = self._namespaced_key(end_key)
|
||||
|
||||
keys = await self.redis.keys(f"{start_key}*")
|
||||
result = []
|
||||
for key in keys:
|
||||
if key <= end_key:
|
||||
value = await self.get(key)
|
||||
if value:
|
||||
result.append(value)
|
||||
return result
|
||||
|
|
@ -1,15 +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.
|
||||
|
||||
from .config import SqliteControlPlaneConfig
|
||||
|
||||
|
||||
async def get_provider_impl(config: SqliteControlPlaneConfig, _deps):
|
||||
from .control_plane import SqliteControlPlane
|
||||
|
||||
impl = SqliteControlPlane(config)
|
||||
await impl.initialize()
|
||||
return impl
|
||||
|
|
@ -1,19 +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.
|
||||
|
||||
from llama_models.schema_utils import json_schema_type
|
||||
from pydantic import BaseModel, Field
|
||||
|
||||
|
||||
@json_schema_type
|
||||
class SqliteControlPlaneConfig(BaseModel):
|
||||
db_path: str = Field(
|
||||
description="File path for the sqlite database",
|
||||
)
|
||||
table_name: str = Field(
|
||||
default="llamastack_control_plane",
|
||||
description="Table into which all the keys will be placed",
|
||||
)
|
||||
|
|
@ -1,79 +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 json
|
||||
from datetime import datetime
|
||||
from typing import Any, List, Optional
|
||||
|
||||
import aiosqlite
|
||||
|
||||
from llama_stack.apis.control_plane import * # noqa: F403
|
||||
|
||||
|
||||
from .config import SqliteControlPlaneConfig
|
||||
|
||||
|
||||
class SqliteControlPlane(ControlPlane):
|
||||
def __init__(self, config: SqliteControlPlaneConfig):
|
||||
self.db_path = config.db_path
|
||||
self.table_name = config.table_name
|
||||
|
||||
async def initialize(self):
|
||||
async with aiosqlite.connect(self.db_path) as db:
|
||||
await db.execute(
|
||||
f"""
|
||||
CREATE TABLE IF NOT EXISTS {self.table_name} (
|
||||
key TEXT PRIMARY KEY,
|
||||
value TEXT,
|
||||
expiration TIMESTAMP
|
||||
)
|
||||
"""
|
||||
)
|
||||
await db.commit()
|
||||
|
||||
async def set(
|
||||
self, key: str, value: Any, expiration: Optional[datetime] = None
|
||||
) -> None:
|
||||
async with aiosqlite.connect(self.db_path) as db:
|
||||
await db.execute(
|
||||
f"INSERT OR REPLACE INTO {self.table_name} (key, value, expiration) VALUES (?, ?, ?)",
|
||||
(key, json.dumps(value), expiration),
|
||||
)
|
||||
await db.commit()
|
||||
|
||||
async def get(self, key: str) -> Optional[ControlPlaneValue]:
|
||||
async with aiosqlite.connect(self.db_path) as db:
|
||||
async with db.execute(
|
||||
f"SELECT value, expiration FROM {self.table_name} WHERE key = ?", (key,)
|
||||
) as cursor:
|
||||
row = await cursor.fetchone()
|
||||
if row is None:
|
||||
return None
|
||||
value, expiration = row
|
||||
return ControlPlaneValue(
|
||||
key=key, value=json.loads(value), expiration=expiration
|
||||
)
|
||||
|
||||
async def delete(self, key: str) -> None:
|
||||
async with aiosqlite.connect(self.db_path) as db:
|
||||
await db.execute(f"DELETE FROM {self.table_name} WHERE key = ?", (key,))
|
||||
await db.commit()
|
||||
|
||||
async def range(self, start_key: str, end_key: str) -> List[ControlPlaneValue]:
|
||||
async with aiosqlite.connect(self.db_path) as db:
|
||||
async with db.execute(
|
||||
f"SELECT key, value, expiration FROM {self.table_name} WHERE key >= ? AND key <= ?",
|
||||
(start_key, end_key),
|
||||
) as cursor:
|
||||
result = []
|
||||
async for row in cursor:
|
||||
key, value, expiration = row
|
||||
result.append(
|
||||
ControlPlaneValue(
|
||||
key=key, value=json.loads(value), expiration=expiration
|
||||
)
|
||||
)
|
||||
return result
|
||||
|
|
@ -1,35 +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.
|
||||
|
||||
from datetime import datetime
|
||||
from typing import Any, List, Optional, Protocol
|
||||
|
||||
from llama_models.schema_utils import json_schema_type, webmethod
|
||||
from pydantic import BaseModel
|
||||
|
||||
|
||||
@json_schema_type
|
||||
class ControlPlaneValue(BaseModel):
|
||||
key: str
|
||||
value: Any
|
||||
expiration: Optional[datetime] = None
|
||||
|
||||
|
||||
@json_schema_type
|
||||
class ControlPlane(Protocol):
|
||||
@webmethod(route="/control_plane/set")
|
||||
async def set(
|
||||
self, key: str, value: Any, expiration: Optional[datetime] = None
|
||||
) -> None: ...
|
||||
|
||||
@webmethod(route="/control_plane/get", method="GET")
|
||||
async def get(self, key: str) -> Optional[ControlPlaneValue]: ...
|
||||
|
||||
@webmethod(route="/control_plane/delete")
|
||||
async def delete(self, key: str) -> None: ...
|
||||
|
||||
@webmethod(route="/control_plane/range", method="GET")
|
||||
async def range(self, start_key: str, end_key: str) -> List[ControlPlaneValue]: ...
|
||||
|
|
@ -1,29 +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.
|
||||
|
||||
from typing import List
|
||||
|
||||
from llama_stack.distribution.datatypes import * # noqa: F403
|
||||
|
||||
|
||||
def available_providers() -> List[ProviderSpec]:
|
||||
return [
|
||||
InlineProviderSpec(
|
||||
api=Api.control_plane,
|
||||
provider_id="sqlite",
|
||||
pip_packages=["aiosqlite"],
|
||||
module="llama_stack.providers.impls.sqlite.control_plane",
|
||||
config_class="llama_stack.providers.impls.sqlite.control_plane.SqliteControlPlaneConfig",
|
||||
),
|
||||
remote_provider_spec(
|
||||
Api.control_plane,
|
||||
AdapterSpec(
|
||||
adapter_id="redis",
|
||||
pip_packages=["redis"],
|
||||
module="llama_stack.providers.adapters.control_plane.redis",
|
||||
),
|
||||
),
|
||||
]
|
||||
Loading…
Add table
Add a link
Reference in a new issue