llama-stack-api 0.4.2__py3-none-any.whl → 0.4.4__py3-none-any.whl

llama_stack_api/common/job_types.py

@@ -0,0 +1,38 @@
+# 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 enum import Enum
+
+from pydantic import BaseModel
+
+from llama_stack_api.schema_utils import json_schema_type
+
+
+class JobStatus(Enum):
+    """Status of a job execution.
+    :cvar completed: Job has finished successfully
+    :cvar in_progress: Job is currently running
+    :cvar failed: Job has failed during execution
+    :cvar scheduled: Job is scheduled but not yet started
+    :cvar cancelled: Job was cancelled before completion
+    """
+
+    completed = "completed"
+    in_progress = "in_progress"
+    failed = "failed"
+    scheduled = "scheduled"
+    cancelled = "cancelled"
+
+
+@json_schema_type
+class Job(BaseModel):
+    """A job execution instance with status tracking.
+
+    :param job_id: Unique identifier for the job
+    :param status: Current execution status of the job
+    """
+
+    job_id: str
+    status: JobStatus

llama_stack_api/common/responses.py

@@ -0,0 +1,77 @@
+# 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 enum import Enum
+from typing import Any
+
+from pydantic import BaseModel
+
+from llama_stack_api.schema_utils import json_schema_type
+
+
+class Order(Enum):
+    """Sort order for paginated responses.
+    :cvar asc: Ascending order
+    :cvar desc: Descending order
+    """
+
+    asc = "asc"
+    desc = "desc"
+
+
+@json_schema_type
+class PaginatedResponse(BaseModel):
+    """A generic paginated response that follows a simple format.
+
+    :param data: The list of items for the current page
+    :param has_more: Whether there are more items available after this set
+    :param url: The URL for accessing this list
+    """
+
+    data: list[dict[str, Any]]
+    has_more: bool
+    url: str | None = None
+
+
+# This is a short term solution to allow inference API to return metrics
+# The ideal way to do this is to have a way for all response types to include metrics
+# and all metric events logged to the telemetry API to be included with the response
+# To do this, we will need to augment all response types with a metrics field.
+# We have hit a blocker from stainless SDK that prevents us from doing this.
+# The blocker is that if we were to augment the response types that have a data field
+# in them like so
+# class ListModelsResponse(BaseModel):
+# metrics: Optional[List[MetricEvent]] = None
+# data: List[Models]
+# ...
+# The client SDK will need to access the data by using a .data field, which is not
+# ergonomic. Stainless SDK does support unwrapping the response type, but it
+# requires that the response type to only have a single field.
+
+# We will need a way in the client SDK to signal that the metrics are needed
+# and if they are needed, the client SDK has to return the full response type
+# without unwrapping it.
+
+
+@json_schema_type
+class MetricInResponse(BaseModel):
+    """A metric value included in API responses.
+    :param metric: The name of the metric
+    :param value: The numeric value of the metric
+    :param unit: (Optional) The unit of measurement for the metric value
+    """
+
+    metric: str
+    value: int | float
+    unit: str | None = None
+
+
+class MetricResponseMixin(BaseModel):
+    """Mixin class for API responses that can include metrics.
+    :param metrics: (Optional) List of metrics associated with the API response
+    """
+
+    metrics: list[MetricInResponse] | None = None

llama_stack_api/common/training_types.py

@@ -0,0 +1,47 @@
+# 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 pydantic import BaseModel
+
+from llama_stack_api.schema_utils import json_schema_type
+
+
+@json_schema_type
+class PostTrainingMetric(BaseModel):
+    """Training metrics captured during post-training jobs.
+
+    :param epoch: Training epoch number
+    :param train_loss: Loss value on the training dataset
+    :param validation_loss: Loss value on the validation dataset
+    :param perplexity: Perplexity metric indicating model confidence
+    """
+
+    epoch: int
+    train_loss: float
+    validation_loss: float
+    perplexity: float
+
+
+@json_schema_type
+class Checkpoint(BaseModel):
+    """Checkpoint created during training runs.
+
+    :param identifier: Unique identifier for the checkpoint
+    :param created_at: Timestamp when the checkpoint was created
+    :param epoch: Training epoch when the checkpoint was saved
+    :param post_training_job_id: Identifier of the training job that created this checkpoint
+    :param path: File system path where the checkpoint is stored
+    :param training_metrics: (Optional) Training metrics associated with this checkpoint
+    """
+
+    identifier: str
+    created_at: datetime
+    epoch: int
+    post_training_job_id: str
+    path: str
+    training_metrics: PostTrainingMetric | None = None

llama_stack_api/common/type_system.py

@@ -0,0 +1,146 @@
+# 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 Annotated, Literal
+
+from pydantic import BaseModel, Field
+
+from llama_stack_api.schema_utils import json_schema_type, register_schema
+
+
+@json_schema_type
+class StringType(BaseModel):
+    """Parameter type for string values.
+
+    :param type: Discriminator type. Always "string"
+    """
+
+    type: Literal["string"] = "string"
+
+
+@json_schema_type
+class NumberType(BaseModel):
+    """Parameter type for numeric values.
+
+    :param type: Discriminator type. Always "number"
+    """
+
+    type: Literal["number"] = "number"
+
+
+@json_schema_type
+class BooleanType(BaseModel):
+    """Parameter type for boolean values.
+
+    :param type: Discriminator type. Always "boolean"
+    """
+
+    type: Literal["boolean"] = "boolean"
+
+
+@json_schema_type
+class ArrayType(BaseModel):
+    """Parameter type for array values.
+
+    :param type: Discriminator type. Always "array"
+    """
+
+    type: Literal["array"] = "array"
+
+
+@json_schema_type
+class ObjectType(BaseModel):
+    """Parameter type for object values.
+
+    :param type: Discriminator type. Always "object"
+    """
+
+    type: Literal["object"] = "object"
+
+
+@json_schema_type
+class JsonType(BaseModel):
+    """Parameter type for JSON values.
+
+    :param type: Discriminator type. Always "json"
+    """
+
+    type: Literal["json"] = "json"
+
+
+@json_schema_type
+class UnionType(BaseModel):
+    """Parameter type for union values.
+
+    :param type: Discriminator type. Always "union"
+    """
+
+    type: Literal["union"] = "union"
+
+
+@json_schema_type
+class ChatCompletionInputType(BaseModel):
+    """Parameter type for chat completion input.
+
+    :param type: Discriminator type. Always "chat_completion_input"
+    """
+
+    # expects List[Message] for messages
+    type: Literal["chat_completion_input"] = "chat_completion_input"
+
+
+@json_schema_type
+class CompletionInputType(BaseModel):
+    """Parameter type for completion input.
+
+    :param type: Discriminator type. Always "completion_input"
+    """
+
+    # expects InterleavedTextMedia for content
+    type: Literal["completion_input"] = "completion_input"
+
+
+@json_schema_type
+class DialogType(BaseModel):
+    """Parameter type for dialog data with semantic output labels.
+
+    :param type: Discriminator type. Always "dialog"
+    """
+
+    # expects List[Message] for messages
+    # this type semantically contains the output label whereas ChatCompletionInputType does not
+    type: Literal["dialog"] = "dialog"
+
+
+ParamType = Annotated[
+    StringType
+    | NumberType
+    | BooleanType
+    | ArrayType
+    | ObjectType
+    | JsonType
+    | UnionType
+    | ChatCompletionInputType
+    | CompletionInputType,
+    Field(discriminator="type"),
+]
+register_schema(ParamType, name="ParamType")
+
+"""
+# TODO: recursive definition of ParamType in these containers
+# will cause infinite recursion in OpenAPI generation script
+# since we are going with ChatCompletionInputType and CompletionInputType
+# we don't need to worry about ArrayType/ObjectType/UnionType for now
+ArrayType.model_rebuild()
+ObjectType.model_rebuild()
+UnionType.model_rebuild()
+
+
+class CustomType(BaseModel):
+pylint: disable=syntax-error
+    type: Literal["custom"] = "custom"
+    validator_class: str
+"""

llama_stack_api/connectors.py

@@ -0,0 +1,146 @@
+# 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 enum import StrEnum
+from typing import Literal, Protocol
+
+from pydantic import BaseModel, Field
+from typing_extensions import runtime_checkable
+
+from llama_stack_api.resource import Resource, ResourceType
+from llama_stack_api.schema_utils import json_schema_type, webmethod
+from llama_stack_api.tools import ToolDef
+from llama_stack_api.version import LLAMA_STACK_API_V1ALPHA
+
+
+@json_schema_type
+class ConnectorType(StrEnum):
+    """Type of connector."""
+
+    MCP = "mcp"
+
+
+class CommonConnectorFields(BaseModel):
+    """Common fields for all connectors.
+
+    :param connector_type: Type of connector
+    :param connector_id: Identifier for the connector
+    :param url: URL of the connector
+    :param server_label: (Optional) Label of the server
+    """
+
+    connector_type: ConnectorType = Field(default=ConnectorType.MCP)
+    connector_id: str = Field(..., description="Identifier for the connector")
+    url: str = Field(..., description="URL of the connector")
+    server_label: str | None = Field(default=None, description="Label of the server")
+
+
+@json_schema_type
+class Connector(CommonConnectorFields, Resource):
+    """A connector resource representing a connector registered in Llama Stack.
+
+    :param type: Type of resource, always 'connector' for connectors
+    :param server_name: (Optional) Name of the server
+    :param server_description: (Optional) Description of the server
+    """
+
+    model_config = {"populate_by_name": True}
+    type: Literal[ResourceType.connector] = ResourceType.connector
+    server_name: str | None = Field(default=None, description="Name of the server")
+    server_description: str | None = Field(default=None, description="Description of the server")
+
+
+@json_schema_type
+class ConnectorInput(CommonConnectorFields):
+    """Input for creating a connector
+
+    :param type: Type of resource, always 'connector' for connectors
+    """
+
+    type: Literal[ResourceType.connector] = ResourceType.connector
+
+
+@json_schema_type
+class ListConnectorsResponse(BaseModel):
+    """Response containing a list of connectors.
+
+    :param data: List of connectors
+    """
+
+    data: list[Connector]
+
+
+@json_schema_type
+class ListToolsResponse(BaseModel):
+    """Response containing a list of tools.
+
+    :param data: List of tools
+    """
+
+    data: list[ToolDef]
+
+
+@runtime_checkable
+class Connectors(Protocol):
+    # NOTE: Route order matters! More specific routes must come before less specific ones.
+    # Routes with {param:path} are greedy and will match everything including slashes.
+
+    @webmethod(route="/connectors", method="GET", level=LLAMA_STACK_API_V1ALPHA)
+    async def list_connectors(
+        self,
+    ) -> ListConnectorsResponse:
+        """List all configured connectors.
+
+        :returns: A ListConnectorsResponse.
+        """
+        ...
+
+    @webmethod(route="/connectors/{connector_id}/tools/{tool_name}", method="GET", level=LLAMA_STACK_API_V1ALPHA)
+    async def get_connector_tool(
+        self,
+        connector_id: str,
+        tool_name: str,
+        authorization: str | None = None,
+    ) -> ToolDef:
+        """Get a tool definition by its name from a connector.
+
+        :param connector_id: The ID of the connector to get the tool from.
+        :param tool_name: The name of the tool to get.
+        :param authorization: (Optional) OAuth access token for authenticating with the MCP server.
+
+        :returns: A ToolDef.
+        """
+        ...
+
+    @webmethod(route="/connectors/{connector_id}/tools", method="GET", level=LLAMA_STACK_API_V1ALPHA)
+    async def list_connector_tools(
+        self,
+        connector_id: str,
+        authorization: str | None = None,
+    ) -> ListToolsResponse:
+        """List tools available from a connector.
+
+        :param connector_id: The ID of the connector to list tools for.
+        :param authorization: (Optional) OAuth access token for authenticating with the MCP server.
+
+        :returns: A ListToolsResponse.
+        """
+        ...
+
+    @webmethod(route="/connectors/{connector_id}", method="GET", level=LLAMA_STACK_API_V1ALPHA)
+    async def get_connector(
+        self,
+        connector_id: str,
+        authorization: str | None = None,
+    ) -> Connector:
+        """Get a connector by its ID.
+
+        :param connector_id: The ID of the connector to get.
+        :param authorization: (Optional) OAuth access token for authenticating with the MCP server.
+
+        :returns: A Connector.
+        """
+        ...