agents.py

# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
from __future__ import annotations

import time

import httpx

from .routes import (
    RoutesResource,
    AsyncRoutesResource,
    RoutesResourceWithRawResponse,
    AsyncRoutesResourceWithRawResponse,
    RoutesResourceWithStreamingResponse,
    AsyncRoutesResourceWithStreamingResponse,
)
from ...types import (
    APIRetrievalMethod,
    APIDeploymentVisibility,
    agent_list_params,
    agent_create_params,
    agent_update_params,
    agent_update_status_params,
    agent_retrieve_usage_params,
)
from ..._types import Body, Omit, Query, Headers, NotGiven, SequenceNotStr, omit, not_given
from ..._utils import maybe_transform, async_maybe_transform
from .api_keys import (
    APIKeysResource,
    AsyncAPIKeysResource,
    APIKeysResourceWithRawResponse,
    AsyncAPIKeysResourceWithRawResponse,
    APIKeysResourceWithStreamingResponse,
    AsyncAPIKeysResourceWithStreamingResponse,
)
from .versions import (
    VersionsResource,
    AsyncVersionsResource,
    VersionsResourceWithRawResponse,
    AsyncVersionsResourceWithRawResponse,
    VersionsResourceWithStreamingResponse,
    AsyncVersionsResourceWithStreamingResponse,
)
from ..._compat import cached_property
from .chat.chat import (
    ChatResource,
    AsyncChatResource,
    ChatResourceWithRawResponse,
    AsyncChatResourceWithRawResponse,
    ChatResourceWithStreamingResponse,
    AsyncChatResourceWithStreamingResponse,
)
from .functions import (
    FunctionsResource,
    AsyncFunctionsResource,
    FunctionsResourceWithRawResponse,
    AsyncFunctionsResourceWithRawResponse,
    FunctionsResourceWithStreamingResponse,
    AsyncFunctionsResourceWithStreamingResponse,
)
from ..._resource import SyncAPIResource, AsyncAPIResource
from ..._response import (
    to_raw_response_wrapper,
    to_streamed_response_wrapper,
    async_to_raw_response_wrapper,
    async_to_streamed_response_wrapper,
)
from ..._base_client import make_request_options
from .evaluation_runs import (
    EvaluationRunsResource,
    AsyncEvaluationRunsResource,
    EvaluationRunsResourceWithRawResponse,
    AsyncEvaluationRunsResourceWithRawResponse,
    EvaluationRunsResourceWithStreamingResponse,
    AsyncEvaluationRunsResourceWithStreamingResponse,
)
from .knowledge_bases import (
    KnowledgeBasesResource,
    AsyncKnowledgeBasesResource,
    KnowledgeBasesResourceWithRawResponse,
    AsyncKnowledgeBasesResourceWithRawResponse,
    KnowledgeBasesResourceWithStreamingResponse,
    AsyncKnowledgeBasesResourceWithStreamingResponse,
)
from .evaluation_datasets import (
    EvaluationDatasetsResource,
    AsyncEvaluationDatasetsResource,
    EvaluationDatasetsResourceWithRawResponse,
    AsyncEvaluationDatasetsResourceWithRawResponse,
    EvaluationDatasetsResourceWithStreamingResponse,
    AsyncEvaluationDatasetsResourceWithStreamingResponse,
)
from .evaluation_test_cases import (
    EvaluationTestCasesResource,
    AsyncEvaluationTestCasesResource,
    EvaluationTestCasesResourceWithRawResponse,
    AsyncEvaluationTestCasesResourceWithRawResponse,
    EvaluationTestCasesResourceWithStreamingResponse,
    AsyncEvaluationTestCasesResourceWithStreamingResponse,
)
from ...types.agent_list_response import AgentListResponse
from ...types.api_retrieval_method import APIRetrievalMethod
from ...types.agent_create_response import AgentCreateResponse
from ...types.agent_delete_response import AgentDeleteResponse
from ...types.agent_update_response import AgentUpdateResponse
from ...types.agent_retrieve_response import AgentRetrieveResponse
from ...types.api_deployment_visibility import APIDeploymentVisibility
from ...types.agent_update_status_response import AgentUpdateStatusResponse
from ...types.agent_retrieve_usage_response import AgentRetrieveUsageResponse
from .evaluation_metrics.evaluation_metrics import (
    EvaluationMetricsResource,
    AsyncEvaluationMetricsResource,
    EvaluationMetricsResourceWithRawResponse,
    AsyncEvaluationMetricsResourceWithRawResponse,
    EvaluationMetricsResourceWithStreamingResponse,
    AsyncEvaluationMetricsResourceWithStreamingResponse,
)

__all__ = ["AgentsResource", "AsyncAgentsResource"]


class AgentsResource(SyncAPIResource):
    """
    The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
    """

    @cached_property
    def api_keys(self) -> APIKeysResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return APIKeysResource(self._client)

    @cached_property
    def chat(self) -> ChatResource:
        return ChatResource(self._client)

    @cached_property
    def evaluation_metrics(self) -> EvaluationMetricsResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationMetricsResource(self._client)

    @cached_property
    def evaluation_runs(self) -> EvaluationRunsResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationRunsResource(self._client)

    @cached_property
    def evaluation_test_cases(self) -> EvaluationTestCasesResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationTestCasesResource(self._client)

    @cached_property
    def evaluation_datasets(self) -> EvaluationDatasetsResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationDatasetsResource(self._client)

    @cached_property
    def functions(self) -> FunctionsResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return FunctionsResource(self._client)

    @cached_property
    def versions(self) -> VersionsResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return VersionsResource(self._client)

    @cached_property
    def knowledge_bases(self) -> KnowledgeBasesResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return KnowledgeBasesResource(self._client)

    @cached_property
    def routes(self) -> RoutesResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return RoutesResource(self._client)

    @cached_property
    def with_raw_response(self) -> AgentsResourceWithRawResponse:
        """
        This property can be used as a prefix for any HTTP method call to return
        the raw response object instead of the parsed content.

        For more information, see https://www.github.com/digitalocean/gradient-python#accessing-raw-response-data-eg-headers
        """
        return AgentsResourceWithRawResponse(self)

    @cached_property
    def with_streaming_response(self) -> AgentsResourceWithStreamingResponse:
        """
        An alternative to `.with_raw_response` that doesn't eagerly read the response body.

        For more information, see https://www.github.com/digitalocean/gradient-python#with_streaming_response
        """
        return AgentsResourceWithStreamingResponse(self)

    def create(
        self,
        *,
        anthropic_key_uuid: str | Omit = omit,
        description: str | Omit = omit,
        instruction: str | Omit = omit,
        knowledge_base_uuid: SequenceNotStr[str] | Omit = omit,
        model_provider_key_uuid: str | Omit = omit,
        model_uuid: str | Omit = omit,
        name: str | Omit = omit,
        openai_key_uuid: str | Omit = omit,
        project_id: str | Omit = omit,
        region: str | Omit = omit,
        tags: SequenceNotStr[str] | Omit = omit,
        workspace_uuid: str | Omit = omit,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentCreateResponse:
        """To create a new agent, send a POST request to `/v2/gen-ai/agents`.

        The response
        body contains a JSON object with the newly created agent object.

        Args:
          anthropic_key_uuid: Optional Anthropic API key ID to use with Anthropic models

          description: A text description of the agent, not used in inference

          instruction: Agent instruction. Instructions help your agent to perform its job effectively.
              See
              [Write Effective Agent Instructions](https://docs.digitalocean.com/products/genai-platform/concepts/best-practices/#agent-instructions)
              for best practices.

          knowledge_base_uuid: Ids of the knowledge base(s) to attach to the agent

          model_uuid: Identifier for the foundation model.

          name: Agent name

          openai_key_uuid: Optional OpenAI API key ID to use with OpenAI models

          project_id: The id of the DigitalOcean project this agent will belong to

          region: The DigitalOcean region to deploy your agent in

          tags: Agent tag to organize related resources

          workspace_uuid: Identifier for the workspace

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        return self._post(
            "/v2/gen-ai/agents"
            if self._client._base_url_overridden
            else "https://api.digitalocean.com/v2/gen-ai/agents",
            body=maybe_transform(
                {
                    "anthropic_key_uuid": anthropic_key_uuid,
                    "description": description,
                    "instruction": instruction,
                    "knowledge_base_uuid": knowledge_base_uuid,
                    "model_provider_key_uuid": model_provider_key_uuid,
                    "model_uuid": model_uuid,
                    "name": name,
                    "openai_key_uuid": openai_key_uuid,
                    "project_id": project_id,
                    "region": region,
                    "tags": tags,
                    "workspace_uuid": workspace_uuid,
                },
                agent_create_params.AgentCreateParams,
            ),
            options=make_request_options(
                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
            ),
            cast_to=AgentCreateResponse,
        )

    def retrieve(
        self,
        uuid: str,
        *,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentRetrieveResponse:
        """To retrieve details of an agent, GET request to `/v2/gen-ai/agents/{uuid}`.

        The
        response body is a JSON object containing the agent.

        Args:
          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        if not uuid:
            raise ValueError(f"Expected a non-empty value for `uuid` but received {uuid!r}")
        return self._get(
            f"/v2/gen-ai/agents/{uuid}"
            if self._client._base_url_overridden
            else f"https://api.digitalocean.com/v2/gen-ai/agents/{uuid}",
            options=make_request_options(
                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
            ),
            cast_to=AgentRetrieveResponse,
        )

    def update(
        self,
        path_uuid: str,
        *,
        agent_log_insights_enabled: bool | Omit = omit,
        allowed_domains: SequenceNotStr[str] | Omit = omit,
        anthropic_key_uuid: str | Omit = omit,
        conversation_logs_enabled: bool | Omit = omit,
        description: str | Omit = omit,
        instruction: str | Omit = omit,
        k: int | Omit = omit,
        max_tokens: int | Omit = omit,
        model_provider_key_uuid: str | Omit = omit,
        model_uuid: str | Omit = omit,
        name: str | Omit = omit,
        openai_key_uuid: str | Omit = omit,
        project_id: str | Omit = omit,
        provide_citations: bool | Omit = omit,
        retrieval_method: APIRetrievalMethod | Omit = omit,
        tags: SequenceNotStr[str] | Omit = omit,
        temperature: float | Omit = omit,
        top_p: float | Omit = omit,
        body_uuid: str | Omit = omit,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentUpdateResponse:
        """To update an agent, send a PUT request to `/v2/gen-ai/agents/{uuid}`.

        The
        response body is a JSON object containing the agent.

        Args:
          allowed_domains: Optional list of allowed domains for the chatbot - Must use fully qualified
              domain name (FQDN) such as https://example.com

          anthropic_key_uuid: Optional anthropic key uuid for use with anthropic models

          conversation_logs_enabled: Optional update of conversation logs enabled

          description: Agent description

          instruction: Agent instruction. Instructions help your agent to perform its job effectively.
              See
              [Write Effective Agent Instructions](https://docs.digitalocean.com/products/genai-platform/concepts/best-practices/#agent-instructions)
              for best practices.

          k: How many results should be considered from an attached knowledge base

          max_tokens: Specifies the maximum number of tokens the model can process in a single input
              or output, set as a number between 1 and 512. This determines the length of each
              response.

          model_provider_key_uuid: Optional Model Provider uuid for use with provider models

          model_uuid: Identifier for the foundation model.

          name: Agent name

          openai_key_uuid: Optional OpenAI key uuid for use with OpenAI models

          project_id: The id of the DigitalOcean project this agent will belong to

          retrieval_method: - RETRIEVAL_METHOD_UNKNOWN: The retrieval method is unknown
              - RETRIEVAL_METHOD_REWRITE: The retrieval method is rewrite
              - RETRIEVAL_METHOD_STEP_BACK: The retrieval method is step back
              - RETRIEVAL_METHOD_SUB_QUERIES: The retrieval method is sub queries
              - RETRIEVAL_METHOD_NONE: The retrieval method is none

          tags: A set of abitrary tags to organize your agent

          temperature: Controls the model’s creativity, specified as a number between 0 and 1. Lower
              values produce more predictable and conservative responses, while higher values
              encourage creativity and variation.

          top_p: Defines the cumulative probability threshold for word selection, specified as a
              number between 0 and 1. Higher values allow for more diverse outputs, while
              lower values ensure focused and coherent responses.

          body_uuid: Unique agent id

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        if not path_uuid:
            raise ValueError(f"Expected a non-empty value for `path_uuid` but received {path_uuid!r}")
        return self._put(
            f"/v2/gen-ai/agents/{path_uuid}"
            if self._client._base_url_overridden
            else f"https://api.digitalocean.com/v2/gen-ai/agents/{path_uuid}",
            body=maybe_transform(
                {
                    "agent_log_insights_enabled": agent_log_insights_enabled,
                    "allowed_domains": allowed_domains,
                    "anthropic_key_uuid": anthropic_key_uuid,
                    "conversation_logs_enabled": conversation_logs_enabled,
                    "description": description,
                    "instruction": instruction,
                    "k": k,
                    "max_tokens": max_tokens,
                    "model_provider_key_uuid": model_provider_key_uuid,
                    "model_uuid": model_uuid,
                    "name": name,
                    "openai_key_uuid": openai_key_uuid,
                    "project_id": project_id,
                    "provide_citations": provide_citations,
                    "retrieval_method": retrieval_method,
                    "tags": tags,
                    "temperature": temperature,
                    "top_p": top_p,
                    "body_uuid": body_uuid,
                },
                agent_update_params.AgentUpdateParams,
            ),
            options=make_request_options(
                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
            ),
            cast_to=AgentUpdateResponse,
        )

    def list(
        self,
        *,
        only_deployed: bool | Omit = omit,
        page: int | Omit = omit,
        per_page: int | Omit = omit,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentListResponse:
        """
        To list all agents, send a GET request to `/v2/gen-ai/agents`.

        Args:
          only_deployed: Only list agents that are deployed.

          page: Page number.

          per_page: Items per page.

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        return self._get(
            "/v2/gen-ai/agents"
            if self._client._base_url_overridden
            else "https://api.digitalocean.com/v2/gen-ai/agents",
            options=make_request_options(
                extra_headers=extra_headers,
                extra_query=extra_query,
                extra_body=extra_body,
                timeout=timeout,
                query=maybe_transform(
                    {
                        "only_deployed": only_deployed,
                        "page": page,
                        "per_page": per_page,
                    },
                    agent_list_params.AgentListParams,
                ),
            ),
            cast_to=AgentListResponse,
        )

    def delete(
        self,
        uuid: str,
        *,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentDeleteResponse:
        """
        To delete an agent, send a DELETE request to `/v2/gen-ai/agents/{uuid}`.

        Args:
          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        if not uuid:
            raise ValueError(f"Expected a non-empty value for `uuid` but received {uuid!r}")
        return self._delete(
            f"/v2/gen-ai/agents/{uuid}"
            if self._client._base_url_overridden
            else f"https://api.digitalocean.com/v2/gen-ai/agents/{uuid}",
            options=make_request_options(
                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
            ),
            cast_to=AgentDeleteResponse,
        )

    def retrieve_usage(
        self,
        uuid: str,
        *,
        start: str | Omit = omit,
        stop: str | Omit = omit,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentRetrieveUsageResponse:
        """
        To get agent usage, send a GET request to `/v2/gen-ai/agents/{uuid}/usage`.
        Returns usage metrics for the specified agent within the provided time range.

        Args:
          start: Return all usage data from this date.

          stop: Return all usage data up to this date, if omitted, will return up to the current
              date.

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        if not uuid:
            raise ValueError(f"Expected a non-empty value for `uuid` but received {uuid!r}")
        return self._get(
            f"/v2/gen-ai/agents/{uuid}/usage"
            if self._client._base_url_overridden
            else f"https://api.digitalocean.com/v2/gen-ai/agents/{uuid}/usage",
            options=make_request_options(
                extra_headers=extra_headers,
                extra_query=extra_query,
                extra_body=extra_body,
                timeout=timeout,
                query=maybe_transform(
                    {
                        "start": start,
                        "stop": stop,
                    },
                    agent_retrieve_usage_params.AgentRetrieveUsageParams,
                ),
            ),
            cast_to=AgentRetrieveUsageResponse,
        )

    def update_status(
        self,
        path_uuid: str,
        *,
        body_uuid: str | Omit = omit,
        visibility: APIDeploymentVisibility | Omit = omit,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentUpdateStatusResponse:
        """Check whether an agent is public or private.

        To update the agent status, send a
        PUT request to `/v2/gen-ai/agents/{uuid}/deployment_visibility`.

        Args:
          body_uuid: Unique id

          visibility: - VISIBILITY_UNKNOWN: The status of the deployment is unknown
              - VISIBILITY_DISABLED: The deployment is disabled and will no longer service
                requests
              - VISIBILITY_PLAYGROUND: Deprecated: No longer a valid state
              - VISIBILITY_PUBLIC: The deployment is public and will service requests from the
                public internet
              - VISIBILITY_PRIVATE: The deployment is private and will only service requests
                from other agents, or through API keys

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        if not path_uuid:
            raise ValueError(f"Expected a non-empty value for `path_uuid` but received {path_uuid!r}")
        return self._put(
            f"/v2/gen-ai/agents/{path_uuid}/deployment_visibility"
            if self._client._base_url_overridden
            else f"https://api.digitalocean.com/v2/gen-ai/agents/{path_uuid}/deployment_visibility",
            body=maybe_transform(
                {
                    "body_uuid": body_uuid,
                    "visibility": visibility,
                },
                agent_update_status_params.AgentUpdateStatusParams,
            ),
            options=make_request_options(
                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
            ),
            cast_to=AgentUpdateStatusResponse,
        )

    def wait_until_ready(
        self,
        uuid: str,
        *,
        timeout: float = 300.0,
        poll_interval: float = 5.0,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
    ) -> AgentRetrieveResponse:
        """Wait for an agent to be ready (deployment status is STATUS_RUNNING).

        This method polls the agent status until it reaches STATUS_RUNNING or a terminal
        error state. It handles timeout and deployment failures automatically.

        Args:
          uuid: The unique identifier of the agent to wait for

          timeout: Maximum time to wait in seconds (default: 300 seconds / 5 minutes)

          poll_interval: Time to wait between status checks in seconds (default: 5 seconds)

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

        Returns:
          AgentRetrieveResponse: The agent response when it reaches STATUS_RUNNING

        Raises:
          AgentDeploymentError: If the agent deployment fails (STATUS_FAILED,
              STATUS_UNDEPLOYMENT_FAILED, or STATUS_DELETED)
          AgentDeploymentTimeoutError: If the agent doesn't reach STATUS_RUNNING
              within the timeout period
          ValueError: If uuid is empty
        """
        from ..._exceptions import AgentDeploymentError, AgentDeploymentTimeoutError

        if not uuid:
            raise ValueError(f"Expected a non-empty value for `uuid` but received {uuid!r}")

        start_time = time.time()

        while True:
            agent_response = self.retrieve(
                uuid,
                extra_headers=extra_headers,
                extra_query=extra_query,
                extra_body=extra_body,
            )

            # Check if agent and deployment exist
            if agent_response.agent and agent_response.agent.deployment:
                status = agent_response.agent.deployment.status

                # Success case
                if status == "STATUS_RUNNING":
                    return agent_response

                # Failure cases
                if status in ("STATUS_FAILED", "STATUS_UNDEPLOYMENT_FAILED", "STATUS_DELETED"):
                    raise AgentDeploymentError(
                        f"Agent deployment failed with status: {status}",
                        status=status,
                    )

            # Check timeout
            elapsed_time = time.time() - start_time
            if elapsed_time >= timeout:
                current_status = (
                    agent_response.agent.deployment.status
                    if agent_response.agent and agent_response.agent.deployment
                    else "UNKNOWN"
                )
                raise AgentDeploymentTimeoutError(
                    f"Agent did not reach STATUS_RUNNING within {timeout} seconds. Current status: {current_status}",
                    agent_id=uuid,
                )

            # Wait before polling again
            time.sleep(poll_interval)


class AsyncAgentsResource(AsyncAPIResource):
    """
    The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
    """

    @cached_property
    def api_keys(self) -> AsyncAPIKeysResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncAPIKeysResource(self._client)

    @cached_property
    def chat(self) -> AsyncChatResource:
        return AsyncChatResource(self._client)

    @cached_property
    def evaluation_metrics(self) -> AsyncEvaluationMetricsResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationMetricsResource(self._client)

    @cached_property
    def evaluation_runs(self) -> AsyncEvaluationRunsResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationRunsResource(self._client)

    @cached_property
    def evaluation_test_cases(self) -> AsyncEvaluationTestCasesResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationTestCasesResource(self._client)

    @cached_property
    def evaluation_datasets(self) -> AsyncEvaluationDatasetsResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationDatasetsResource(self._client)

    @cached_property
    def functions(self) -> AsyncFunctionsResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncFunctionsResource(self._client)

    @cached_property
    def versions(self) -> AsyncVersionsResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncVersionsResource(self._client)

    @cached_property
    def knowledge_bases(self) -> AsyncKnowledgeBasesResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncKnowledgeBasesResource(self._client)

    @cached_property
    def routes(self) -> AsyncRoutesResource:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncRoutesResource(self._client)

    @cached_property
    def with_raw_response(self) -> AsyncAgentsResourceWithRawResponse:
        """
        This property can be used as a prefix for any HTTP method call to return
        the raw response object instead of the parsed content.

        For more information, see https://www.github.com/digitalocean/gradient-python#accessing-raw-response-data-eg-headers
        """
        return AsyncAgentsResourceWithRawResponse(self)

    @cached_property
    def with_streaming_response(self) -> AsyncAgentsResourceWithStreamingResponse:
        """
        An alternative to `.with_raw_response` that doesn't eagerly read the response body.

        For more information, see https://www.github.com/digitalocean/gradient-python#with_streaming_response
        """
        return AsyncAgentsResourceWithStreamingResponse(self)

    async def create(
        self,
        *,
        anthropic_key_uuid: str | Omit = omit,
        description: str | Omit = omit,
        instruction: str | Omit = omit,
        knowledge_base_uuid: SequenceNotStr[str] | Omit = omit,
        model_provider_key_uuid: str | Omit = omit,
        model_uuid: str | Omit = omit,
        name: str | Omit = omit,
        openai_key_uuid: str | Omit = omit,
        project_id: str | Omit = omit,
        region: str | Omit = omit,
        tags: SequenceNotStr[str] | Omit = omit,
        workspace_uuid: str | Omit = omit,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentCreateResponse:
        """To create a new agent, send a POST request to `/v2/gen-ai/agents`.

        The response
        body contains a JSON object with the newly created agent object.

        Args:
          anthropic_key_uuid: Optional Anthropic API key ID to use with Anthropic models

          description: A text description of the agent, not used in inference

          instruction: Agent instruction. Instructions help your agent to perform its job effectively.
              See
              [Write Effective Agent Instructions](https://docs.digitalocean.com/products/genai-platform/concepts/best-practices/#agent-instructions)
              for best practices.

          knowledge_base_uuid: Ids of the knowledge base(s) to attach to the agent

          model_uuid: Identifier for the foundation model.

          name: Agent name

          openai_key_uuid: Optional OpenAI API key ID to use with OpenAI models

          project_id: The id of the DigitalOcean project this agent will belong to

          region: The DigitalOcean region to deploy your agent in

          tags: Agent tag to organize related resources

          workspace_uuid: Identifier for the workspace

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        return await self._post(
            "/v2/gen-ai/agents"
            if self._client._base_url_overridden
            else "https://api.digitalocean.com/v2/gen-ai/agents",
            body=await async_maybe_transform(
                {
                    "anthropic_key_uuid": anthropic_key_uuid,
                    "description": description,
                    "instruction": instruction,
                    "knowledge_base_uuid": knowledge_base_uuid,
                    "model_provider_key_uuid": model_provider_key_uuid,
                    "model_uuid": model_uuid,
                    "name": name,
                    "openai_key_uuid": openai_key_uuid,
                    "project_id": project_id,
                    "region": region,
                    "tags": tags,
                    "workspace_uuid": workspace_uuid,
                },
                agent_create_params.AgentCreateParams,
            ),
            options=make_request_options(
                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
            ),
            cast_to=AgentCreateResponse,
        )

    async def retrieve(
        self,
        uuid: str,
        *,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentRetrieveResponse:
        """To retrieve details of an agent, GET request to `/v2/gen-ai/agents/{uuid}`.

        The
        response body is a JSON object containing the agent.

        Args:
          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        if not uuid:
            raise ValueError(f"Expected a non-empty value for `uuid` but received {uuid!r}")
        return await self._get(
            f"/v2/gen-ai/agents/{uuid}"
            if self._client._base_url_overridden
            else f"https://api.digitalocean.com/v2/gen-ai/agents/{uuid}",
            options=make_request_options(
                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
            ),
            cast_to=AgentRetrieveResponse,
        )

    async def update(
        self,
        path_uuid: str,
        *,
        agent_log_insights_enabled: bool | Omit = omit,
        allowed_domains: SequenceNotStr[str] | Omit = omit,
        anthropic_key_uuid: str | Omit = omit,
        conversation_logs_enabled: bool | Omit = omit,
        description: str | Omit = omit,
        instruction: str | Omit = omit,
        k: int | Omit = omit,
        max_tokens: int | Omit = omit,
        model_provider_key_uuid: str | Omit = omit,
        model_uuid: str | Omit = omit,
        name: str | Omit = omit,
        openai_key_uuid: str | Omit = omit,
        project_id: str | Omit = omit,
        provide_citations: bool | Omit = omit,
        retrieval_method: APIRetrievalMethod | Omit = omit,
        tags: SequenceNotStr[str] | Omit = omit,
        temperature: float | Omit = omit,
        top_p: float | Omit = omit,
        body_uuid: str | Omit = omit,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentUpdateResponse:
        """To update an agent, send a PUT request to `/v2/gen-ai/agents/{uuid}`.

        The
        response body is a JSON object containing the agent.

        Args:
          allowed_domains: Optional list of allowed domains for the chatbot - Must use fully qualified
              domain name (FQDN) such as https://example.com

          anthropic_key_uuid: Optional anthropic key uuid for use with anthropic models

          conversation_logs_enabled: Optional update of conversation logs enabled

          description: Agent description

          instruction: Agent instruction. Instructions help your agent to perform its job effectively.
              See
              [Write Effective Agent Instructions](https://docs.digitalocean.com/products/genai-platform/concepts/best-practices/#agent-instructions)
              for best practices.

          k: How many results should be considered from an attached knowledge base

          max_tokens: Specifies the maximum number of tokens the model can process in a single input
              or output, set as a number between 1 and 512. This determines the length of each
              response.

          model_provider_key_uuid: Optional Model Provider uuid for use with provider models

          model_uuid: Identifier for the foundation model.

          name: Agent name

          openai_key_uuid: Optional OpenAI key uuid for use with OpenAI models

          project_id: The id of the DigitalOcean project this agent will belong to

          retrieval_method: - RETRIEVAL_METHOD_UNKNOWN: The retrieval method is unknown
              - RETRIEVAL_METHOD_REWRITE: The retrieval method is rewrite
              - RETRIEVAL_METHOD_STEP_BACK: The retrieval method is step back
              - RETRIEVAL_METHOD_SUB_QUERIES: The retrieval method is sub queries
              - RETRIEVAL_METHOD_NONE: The retrieval method is none

          tags: A set of abitrary tags to organize your agent

          temperature: Controls the model’s creativity, specified as a number between 0 and 1. Lower
              values produce more predictable and conservative responses, while higher values
              encourage creativity and variation.

          top_p: Defines the cumulative probability threshold for word selection, specified as a
              number between 0 and 1. Higher values allow for more diverse outputs, while
              lower values ensure focused and coherent responses.

          body_uuid: Unique agent id

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        if not path_uuid:
            raise ValueError(f"Expected a non-empty value for `path_uuid` but received {path_uuid!r}")
        return await self._put(
            f"/v2/gen-ai/agents/{path_uuid}"
            if self._client._base_url_overridden
            else f"https://api.digitalocean.com/v2/gen-ai/agents/{path_uuid}",
            body=await async_maybe_transform(
                {
                    "agent_log_insights_enabled": agent_log_insights_enabled,
                    "allowed_domains": allowed_domains,
                    "anthropic_key_uuid": anthropic_key_uuid,
                    "conversation_logs_enabled": conversation_logs_enabled,
                    "description": description,
                    "instruction": instruction,
                    "k": k,
                    "max_tokens": max_tokens,
                    "model_provider_key_uuid": model_provider_key_uuid,
                    "model_uuid": model_uuid,
                    "name": name,
                    "openai_key_uuid": openai_key_uuid,
                    "project_id": project_id,
                    "provide_citations": provide_citations,
                    "retrieval_method": retrieval_method,
                    "tags": tags,
                    "temperature": temperature,
                    "top_p": top_p,
                    "body_uuid": body_uuid,
                },
                agent_update_params.AgentUpdateParams,
            ),
            options=make_request_options(
                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
            ),
            cast_to=AgentUpdateResponse,
        )

    async def list(
        self,
        *,
        only_deployed: bool | Omit = omit,
        page: int | Omit = omit,
        per_page: int | Omit = omit,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentListResponse:
        """
        To list all agents, send a GET request to `/v2/gen-ai/agents`.

        Args:
          only_deployed: Only list agents that are deployed.

          page: Page number.

          per_page: Items per page.

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        return await self._get(
            "/v2/gen-ai/agents"
            if self._client._base_url_overridden
            else "https://api.digitalocean.com/v2/gen-ai/agents",
            options=make_request_options(
                extra_headers=extra_headers,
                extra_query=extra_query,
                extra_body=extra_body,
                timeout=timeout,
                query=await async_maybe_transform(
                    {
                        "only_deployed": only_deployed,
                        "page": page,
                        "per_page": per_page,
                    },
                    agent_list_params.AgentListParams,
                ),
            ),
            cast_to=AgentListResponse,
        )

    async def delete(
        self,
        uuid: str,
        *,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentDeleteResponse:
        """
        To delete an agent, send a DELETE request to `/v2/gen-ai/agents/{uuid}`.

        Args:
          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        if not uuid:
            raise ValueError(f"Expected a non-empty value for `uuid` but received {uuid!r}")
        return await self._delete(
            f"/v2/gen-ai/agents/{uuid}"
            if self._client._base_url_overridden
            else f"https://api.digitalocean.com/v2/gen-ai/agents/{uuid}",
            options=make_request_options(
                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
            ),
            cast_to=AgentDeleteResponse,
        )

    async def retrieve_usage(
        self,
        uuid: str,
        *,
        start: str | Omit = omit,
        stop: str | Omit = omit,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentRetrieveUsageResponse:
        """
        To get agent usage, send a GET request to `/v2/gen-ai/agents/{uuid}/usage`.
        Returns usage metrics for the specified agent within the provided time range.

        Args:
          start: Return all usage data from this date.

          stop: Return all usage data up to this date, if omitted, will return up to the current
              date.

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        if not uuid:
            raise ValueError(f"Expected a non-empty value for `uuid` but received {uuid!r}")
        return await self._get(
            f"/v2/gen-ai/agents/{uuid}/usage"
            if self._client._base_url_overridden
            else f"https://api.digitalocean.com/v2/gen-ai/agents/{uuid}/usage",
            options=make_request_options(
                extra_headers=extra_headers,
                extra_query=extra_query,
                extra_body=extra_body,
                timeout=timeout,
                query=await async_maybe_transform(
                    {
                        "start": start,
                        "stop": stop,
                    },
                    agent_retrieve_usage_params.AgentRetrieveUsageParams,
                ),
            ),
            cast_to=AgentRetrieveUsageResponse,
        )

    async def update_status(
        self,
        path_uuid: str,
        *,
        body_uuid: str | Omit = omit,
        visibility: APIDeploymentVisibility | Omit = omit,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
        timeout: float | httpx.Timeout | None | NotGiven = not_given,
    ) -> AgentUpdateStatusResponse:
        """Check whether an agent is public or private.

        To update the agent status, send a
        PUT request to `/v2/gen-ai/agents/{uuid}/deployment_visibility`.

        Args:
          body_uuid: Unique id

          visibility: - VISIBILITY_UNKNOWN: The status of the deployment is unknown
              - VISIBILITY_DISABLED: The deployment is disabled and will no longer service
                requests
              - VISIBILITY_PLAYGROUND: Deprecated: No longer a valid state
              - VISIBILITY_PUBLIC: The deployment is public and will service requests from the
                public internet
              - VISIBILITY_PRIVATE: The deployment is private and will only service requests
                from other agents, or through API keys

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        if not path_uuid:
            raise ValueError(f"Expected a non-empty value for `path_uuid` but received {path_uuid!r}")
        return await self._put(
            f"/v2/gen-ai/agents/{path_uuid}/deployment_visibility"
            if self._client._base_url_overridden
            else f"https://api.digitalocean.com/v2/gen-ai/agents/{path_uuid}/deployment_visibility",
            body=await async_maybe_transform(
                {
                    "body_uuid": body_uuid,
                    "visibility": visibility,
                },
                agent_update_status_params.AgentUpdateStatusParams,
            ),
            options=make_request_options(
                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
            ),
            cast_to=AgentUpdateStatusResponse,
        )

    async def wait_until_ready(
        self,
        uuid: str,
        *,
        timeout: float = 300.0,
        poll_interval: float = 5.0,
        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
        # The extra values given here take precedence over values defined on the client or passed to this method.
        extra_headers: Headers | None = None,
        extra_query: Query | None = None,
        extra_body: Body | None = None,
    ) -> AgentRetrieveResponse:
        """Wait for an agent to be ready (deployment status is STATUS_RUNNING).

        This method polls the agent status until it reaches STATUS_RUNNING or a terminal
        error state. It handles timeout and deployment failures automatically.

        Args:
          uuid: The unique identifier of the agent to wait for

          timeout: Maximum time to wait in seconds (default: 300 seconds / 5 minutes)

          poll_interval: Time to wait between status checks in seconds (default: 5 seconds)

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

        Returns:
          AgentRetrieveResponse: The agent response when it reaches STATUS_RUNNING

        Raises:
          AgentDeploymentError: If the agent deployment fails (STATUS_FAILED,
              STATUS_UNDEPLOYMENT_FAILED, or STATUS_DELETED)
          AgentDeploymentTimeoutError: If the agent doesn't reach STATUS_RUNNING
              within the timeout period
          ValueError: If uuid is empty
        """
        import asyncio

        from ..._exceptions import AgentDeploymentError, AgentDeploymentTimeoutError

        if not uuid:
            raise ValueError(f"Expected a non-empty value for `uuid` but received {uuid!r}")

        start_time = time.time()

        while True:
            agent_response = await self.retrieve(
                uuid,
                extra_headers=extra_headers,
                extra_query=extra_query,
                extra_body=extra_body,
            )

            # Check if agent and deployment exist
            if agent_response.agent and agent_response.agent.deployment:
                status = agent_response.agent.deployment.status

                # Success case
                if status == "STATUS_RUNNING":
                    return agent_response

                # Failure cases
                if status in ("STATUS_FAILED", "STATUS_UNDEPLOYMENT_FAILED", "STATUS_DELETED"):
                    raise AgentDeploymentError(
                        f"Agent deployment failed with status: {status}",
                        status=status,
                    )

            # Check timeout
            elapsed_time = time.time() - start_time
            if elapsed_time >= timeout:
                current_status = (
                    agent_response.agent.deployment.status
                    if agent_response.agent and agent_response.agent.deployment
                    else "UNKNOWN"
                )
                raise AgentDeploymentTimeoutError(
                    f"Agent did not reach STATUS_RUNNING within {timeout} seconds. Current status: {current_status}",
                    agent_id=uuid,
                )

            # Wait before polling again
            await asyncio.sleep(poll_interval)


class AgentsResourceWithRawResponse:
    def __init__(self, agents: AgentsResource) -> None:
        self._agents = agents

        self.create = to_raw_response_wrapper(
            agents.create,
        )
        self.retrieve = to_raw_response_wrapper(
            agents.retrieve,
        )
        self.update = to_raw_response_wrapper(
            agents.update,
        )
        self.list = to_raw_response_wrapper(
            agents.list,
        )
        self.delete = to_raw_response_wrapper(
            agents.delete,
        )
        self.retrieve_usage = to_raw_response_wrapper(
            agents.retrieve_usage,
        )
        self.update_status = to_raw_response_wrapper(
            agents.update_status,
        )
        self.wait_until_ready = to_raw_response_wrapper(
            agents.wait_until_ready,
        )

    @cached_property
    def api_keys(self) -> APIKeysResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return APIKeysResourceWithRawResponse(self._agents.api_keys)

    @cached_property
    def chat(self) -> ChatResourceWithRawResponse:
        return ChatResourceWithRawResponse(self._agents.chat)

    @cached_property
    def evaluation_metrics(self) -> EvaluationMetricsResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationMetricsResourceWithRawResponse(self._agents.evaluation_metrics)

    @cached_property
    def evaluation_runs(self) -> EvaluationRunsResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationRunsResourceWithRawResponse(self._agents.evaluation_runs)

    @cached_property
    def evaluation_test_cases(self) -> EvaluationTestCasesResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationTestCasesResourceWithRawResponse(self._agents.evaluation_test_cases)

    @cached_property
    def evaluation_datasets(self) -> EvaluationDatasetsResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationDatasetsResourceWithRawResponse(self._agents.evaluation_datasets)

    @cached_property
    def functions(self) -> FunctionsResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return FunctionsResourceWithRawResponse(self._agents.functions)

    @cached_property
    def versions(self) -> VersionsResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return VersionsResourceWithRawResponse(self._agents.versions)

    @cached_property
    def knowledge_bases(self) -> KnowledgeBasesResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return KnowledgeBasesResourceWithRawResponse(self._agents.knowledge_bases)

    @cached_property
    def routes(self) -> RoutesResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return RoutesResourceWithRawResponse(self._agents.routes)


class AsyncAgentsResourceWithRawResponse:
    def __init__(self, agents: AsyncAgentsResource) -> None:
        self._agents = agents

        self.create = async_to_raw_response_wrapper(
            agents.create,
        )
        self.retrieve = async_to_raw_response_wrapper(
            agents.retrieve,
        )
        self.update = async_to_raw_response_wrapper(
            agents.update,
        )
        self.list = async_to_raw_response_wrapper(
            agents.list,
        )
        self.delete = async_to_raw_response_wrapper(
            agents.delete,
        )
        self.retrieve_usage = async_to_raw_response_wrapper(
            agents.retrieve_usage,
        )
        self.update_status = async_to_raw_response_wrapper(
            agents.update_status,
        )
        self.wait_until_ready = async_to_raw_response_wrapper(
            agents.wait_until_ready,
        )

    @cached_property
    def api_keys(self) -> AsyncAPIKeysResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncAPIKeysResourceWithRawResponse(self._agents.api_keys)

    @cached_property
    def chat(self) -> AsyncChatResourceWithRawResponse:
        return AsyncChatResourceWithRawResponse(self._agents.chat)

    @cached_property
    def evaluation_metrics(self) -> AsyncEvaluationMetricsResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationMetricsResourceWithRawResponse(self._agents.evaluation_metrics)

    @cached_property
    def evaluation_runs(self) -> AsyncEvaluationRunsResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationRunsResourceWithRawResponse(self._agents.evaluation_runs)

    @cached_property
    def evaluation_test_cases(self) -> AsyncEvaluationTestCasesResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationTestCasesResourceWithRawResponse(self._agents.evaluation_test_cases)

    @cached_property
    def evaluation_datasets(self) -> AsyncEvaluationDatasetsResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationDatasetsResourceWithRawResponse(self._agents.evaluation_datasets)

    @cached_property
    def functions(self) -> AsyncFunctionsResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncFunctionsResourceWithRawResponse(self._agents.functions)

    @cached_property
    def versions(self) -> AsyncVersionsResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncVersionsResourceWithRawResponse(self._agents.versions)

    @cached_property
    def knowledge_bases(self) -> AsyncKnowledgeBasesResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncKnowledgeBasesResourceWithRawResponse(self._agents.knowledge_bases)

    @cached_property
    def routes(self) -> AsyncRoutesResourceWithRawResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncRoutesResourceWithRawResponse(self._agents.routes)


class AgentsResourceWithStreamingResponse:
    def __init__(self, agents: AgentsResource) -> None:
        self._agents = agents

        self.create = to_streamed_response_wrapper(
            agents.create,
        )
        self.retrieve = to_streamed_response_wrapper(
            agents.retrieve,
        )
        self.update = to_streamed_response_wrapper(
            agents.update,
        )
        self.list = to_streamed_response_wrapper(
            agents.list,
        )
        self.delete = to_streamed_response_wrapper(
            agents.delete,
        )
        self.retrieve_usage = to_streamed_response_wrapper(
            agents.retrieve_usage,
        )
        self.update_status = to_streamed_response_wrapper(
            agents.update_status,
        )
        self.wait_until_ready = to_streamed_response_wrapper(
            agents.wait_until_ready,
        )

    @cached_property
    def api_keys(self) -> APIKeysResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return APIKeysResourceWithStreamingResponse(self._agents.api_keys)

    @cached_property
    def chat(self) -> ChatResourceWithStreamingResponse:
        return ChatResourceWithStreamingResponse(self._agents.chat)

    @cached_property
    def evaluation_metrics(self) -> EvaluationMetricsResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationMetricsResourceWithStreamingResponse(self._agents.evaluation_metrics)

    @cached_property
    def evaluation_runs(self) -> EvaluationRunsResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationRunsResourceWithStreamingResponse(self._agents.evaluation_runs)

    @cached_property
    def evaluation_test_cases(self) -> EvaluationTestCasesResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationTestCasesResourceWithStreamingResponse(self._agents.evaluation_test_cases)

    @cached_property
    def evaluation_datasets(self) -> EvaluationDatasetsResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return EvaluationDatasetsResourceWithStreamingResponse(self._agents.evaluation_datasets)

    @cached_property
    def functions(self) -> FunctionsResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return FunctionsResourceWithStreamingResponse(self._agents.functions)

    @cached_property
    def versions(self) -> VersionsResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return VersionsResourceWithStreamingResponse(self._agents.versions)

    @cached_property
    def knowledge_bases(self) -> KnowledgeBasesResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return KnowledgeBasesResourceWithStreamingResponse(self._agents.knowledge_bases)

    @cached_property
    def routes(self) -> RoutesResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return RoutesResourceWithStreamingResponse(self._agents.routes)


class AsyncAgentsResourceWithStreamingResponse:
    def __init__(self, agents: AsyncAgentsResource) -> None:
        self._agents = agents

        self.create = async_to_streamed_response_wrapper(
            agents.create,
        )
        self.retrieve = async_to_streamed_response_wrapper(
            agents.retrieve,
        )
        self.update = async_to_streamed_response_wrapper(
            agents.update,
        )
        self.list = async_to_streamed_response_wrapper(
            agents.list,
        )
        self.delete = async_to_streamed_response_wrapper(
            agents.delete,
        )
        self.retrieve_usage = async_to_streamed_response_wrapper(
            agents.retrieve_usage,
        )
        self.update_status = async_to_streamed_response_wrapper(
            agents.update_status,
        )
        self.wait_until_ready = async_to_streamed_response_wrapper(
            agents.wait_until_ready,
        )

    @cached_property
    def api_keys(self) -> AsyncAPIKeysResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncAPIKeysResourceWithStreamingResponse(self._agents.api_keys)

    @cached_property
    def chat(self) -> AsyncChatResourceWithStreamingResponse:
        return AsyncChatResourceWithStreamingResponse(self._agents.chat)

    @cached_property
    def evaluation_metrics(self) -> AsyncEvaluationMetricsResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationMetricsResourceWithStreamingResponse(self._agents.evaluation_metrics)

    @cached_property
    def evaluation_runs(self) -> AsyncEvaluationRunsResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationRunsResourceWithStreamingResponse(self._agents.evaluation_runs)

    @cached_property
    def evaluation_test_cases(self) -> AsyncEvaluationTestCasesResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationTestCasesResourceWithStreamingResponse(self._agents.evaluation_test_cases)

    @cached_property
    def evaluation_datasets(self) -> AsyncEvaluationDatasetsResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncEvaluationDatasetsResourceWithStreamingResponse(self._agents.evaluation_datasets)

    @cached_property
    def functions(self) -> AsyncFunctionsResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncFunctionsResourceWithStreamingResponse(self._agents.functions)

    @cached_property
    def versions(self) -> AsyncVersionsResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncVersionsResourceWithStreamingResponse(self._agents.versions)

    @cached_property
    def knowledge_bases(self) -> AsyncKnowledgeBasesResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncKnowledgeBasesResourceWithStreamingResponse(self._agents.knowledge_bases)

    @cached_property
    def routes(self) -> AsyncRoutesResourceWithStreamingResponse:
        """
        The API lets you build GPU-powered AI agents with pre-built or custom foundation models, function and agent routes, and RAG pipelines with knowledge bases.
        """
        return AsyncRoutesResourceWithStreamingResponse(self._agents.routes)