phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-09 19:29:18 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Another round of simplification and clarity for models/shields/memory_banks stuff

Browse source
This commit is contained in:
Ashwin Bharambe 2024-10-09 19:19:26 -07:00
parent 73a0a34e39
commit b55034c0de
27 changed files with 454 additions and 444 deletions
Download patch file Download diff file Expand all files Collapse all files

29
llama_stack/apis/models/models.py
View file

@ -4,34 +4,39 @@
# 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, Optional, Protocol
from typing import Any, Dict, List, Optional, Protocol, runtime_checkable
from llama_models.schema_utils import json_schema_type, webmethod
from pydantic import BaseModel, Field
@json_schema_type
class ModelDef(BaseModel):
identifier: str = Field(
description="A unique identifier for the model type",
description="A unique name for the model type",
)
llama_model: str = Field(
description="Pointer to the core Llama family model",
description="Pointer to the underlying core Llama family model. Each model served by Llama Stack must have a core Llama model.",
)
provider_id: Optional[str] = Field(
default=None, description="The provider instance which serves this model"
metadata: Dict[str, Any] = Field(
default_factory=dict,
description="Any additional metadata for this model",
)
# For now, we are only supporting core llama models but as soon as finetuned
# and other custom models (for example various quantizations) are allowed, there
# will be more metadata fields here
@json_schema_type
class ModelDefWithProvider(ModelDef):
provider_id: str = Field(
description="The provider ID for this model",
)
@runtime_checkable
class Models(Protocol):
@webmethod(route="/models/list", method="GET")
async def list_models(self) -> List[ModelDef]: ...
async def list_models(self) -> List[ModelDefWithProvider]: ...
@webmethod(route="/models/get", method="GET")
async def get_model(self, identifier: str) -> Optional[ModelDef]: ...
async def get_model(self, identifier: str) -> Optional[ModelDefWithProvider]: ...
@webmethod(route="/models/register", method="POST")
async def register_model(self, model: ModelDef) -> None: ...
async def register_model(self, model: ModelDefWithProvider) -> None: ...