ibm-watsonx-orchestrate 1.5.0b0__py3-none-any.whl → 1.5.0b1__py3-none-any.whl

ibm_watsonx_orchestrate/cli/commands/models/models_controller.py

@@ -0,0 +1,437 @@
+import logging
+import os
+import sys
+import json
+import yaml
+import importlib
+import inspect
+from pathlib import Path
+from typing import List
+
+import requests
+import rich
+import rich.highlighter
+
+from ibm_watsonx_orchestrate.cli.commands.server.server_command import get_default_env_file, merge_env
+from ibm_watsonx_orchestrate.client.model_policies.model_policies_client import ModelPoliciesClient
+from ibm_watsonx_orchestrate.agent_builder.model_policies.types import ModelPolicy, ModelPolicyInner, \
+    ModelPolicyRetry, ModelPolicyStrategy, ModelPolicyStrategyMode, ModelPolicyTarget
+from ibm_watsonx_orchestrate.client.models.models_client import ModelsClient
+from ibm_watsonx_orchestrate.agent_builder.models.types import VirtualModel, ProviderConfig, ModelType, ANTHROPIC_DEFAULT_MAX_TOKENS
+from ibm_watsonx_orchestrate.client.utils import instantiate_client
+from ibm_watsonx_orchestrate.client.connections import get_connection_id, ConnectionType
+
+logger = logging.getLogger(__name__)
+
+WATSONX_URL = os.getenv("WATSONX_URL")
+
+class ModelHighlighter(rich.highlighter.RegexHighlighter):
+    base_style = "model."
+    highlights = [r"(?P<name>(watsonx|virtual[-]model|virtual[-]policy)\/.+\/.+):"]
+
+def _get_wxai_foundational_models() -> dict:
+    foundation_models_url = WATSONX_URL + "/ml/v1/foundation_model_specs?version=2024-05-01"
+
+    try:
+        response = requests.get(foundation_models_url)
+    except requests.exceptions.RequestException as e:
+        logger.exception(f"Exception when connecting to Watsonx URL: {foundation_models_url}")
+        raise
+
+    if response.status_code != 200:
+        error_message = (
+            f"Failed to retrieve foundational models from {foundation_models_url}. "
+            f"Status code: {response.status_code}. Response: {response.content}"
+        )
+        raise Exception(error_message)
+    
+    json_response = response.json()
+    return json_response
+
+def _string_to_list(env_value) -> List[str]:
+    return [item.strip().lower() for item in env_value.split(",") if item.strip()]
+
+def create_model_from_spec(spec: dict) -> VirtualModel:
+    return VirtualModel.model_validate(spec)
+
+def create_policy_from_spec(spec: dict) -> ModelPolicy:
+    return ModelPolicy.model_validate(spec)
+
+def import_python_model(file: str) -> List[VirtualModel]:
+    file_path = Path(file)
+    file_directory = file_path.parent
+    file_name = file_path.stem
+    sys.path.append(str(file_directory))
+    module = importlib.import_module(file_name)
+    del sys.path[-1]
+
+    models = []
+    for _, obj in inspect.getmembers(module):
+        if isinstance(obj, VirtualModel):
+            models.append(obj)
+    return models
+
+def import_python_policy(file: str) -> List[ModelPolicy]:
+    file_path = Path(file)
+    file_directory = file_path.parent
+    file_name = file_path.stem
+    sys.path.append(str(file_directory))
+    module = importlib.import_module(file_name)
+    del sys.path[-1]
+
+    models = []
+    for _, obj in inspect.getmembers(module):
+        if isinstance(obj, ModelPolicy):
+            models.append(obj)
+    return models
+
+def parse_model_file(file: str) -> List[VirtualModel]:
+    if file.endswith('.yaml') or file.endswith('.yml') or file.endswith(".json"):
+        with open(file, 'r') as f:
+            if file.endswith(".json"):
+                content = json.load(f)
+            else:
+                content = yaml.load(f, Loader=yaml.SafeLoader)
+        model = create_model_from_spec(spec=content)
+        return [model]
+    elif file.endswith('.py'):
+        models = import_python_model(file)
+        return models
+    else:
+        raise ValueError("file must end in .json, .yaml, .yml or .py")
+
+def parse_policy_file(file: str) -> List[ModelPolicy]:
+    if file.endswith('.yaml') or file.endswith('.yml') or file.endswith(".json"):
+        with open(file, 'r') as f:
+            if file.endswith(".json"):
+                content = json.load(f)
+            else:
+                content = yaml.load(f, Loader=yaml.SafeLoader)
+        policy = create_policy_from_spec(spec=content)
+        return [policy]
+    elif file.endswith('.py'):
+        policies = import_python_policy(file)
+        return policies
+    else:
+        raise ValueError("file must end in .json, .yaml, .yml or .py")
+
+def extract_model_names_from_policy_inner(policy_inner: ModelPolicyInner) -> List[str]:
+    model_names = []
+    for target in policy_inner.targets:
+        if isinstance(target, ModelPolicyTarget):
+            model_names.append(target.model_name)
+        elif isinstance(target, ModelPolicyInner):
+            model_names += extract_model_names_from_policy_inner(target)
+    return model_names
+
+def get_model_names_from_policy(policy: ModelPolicy) -> List[str]:
+    return extract_model_names_from_policy_inner(policy_inner=policy.policy)
+
+class ModelsController:
+    def __init__(self):
+        self.models_client = None
+        self.model_policies_client = None
+
+    def get_models_client(self) -> ModelsClient:
+        if not self.models_client:
+            self.models_client = instantiate_client(ModelsClient)
+        return self.models_client
+
+    def get_model_policies_client(self) -> ModelPoliciesClient:
+        if not self.model_policies_client:
+            self.model_policies_client = instantiate_client(ModelPoliciesClient)
+        return self.model_policies_client
+
+    def list_models(self, print_raw: bool = False) -> None:
+        models_client: ModelsClient = self.get_models_client()
+        model_policies_client: ModelPoliciesClient = self.get_model_policies_client()
+        global WATSONX_URL
+        default_env_path = get_default_env_file()
+        merged_env_dict = merge_env(
+            default_env_path,
+            None
+        )
+
+        if 'WATSONX_URL' in merged_env_dict and merged_env_dict['WATSONX_URL']:
+            WATSONX_URL = merged_env_dict['WATSONX_URL']
+
+        watsonx_url = merged_env_dict.get("WATSONX_URL")
+        if not watsonx_url:
+            logger.error("Error: WATSONX_URL is required in the environment.")
+            sys.exit(1)
+
+        logger.info("Retrieving virtual-model models list...")
+        virtual_models = models_client.list()
+
+
+
+        logger.info("Retrieving virtual-policies models list...")
+        virtual_model_policies = model_policies_client.list()
+
+        logger.info("Retrieving watsonx.ai models list...")
+        found_models = _get_wxai_foundational_models()
+
+
+        preferred_str = merged_env_dict.get('PREFERRED_MODELS', '')
+        incompatible_str = merged_env_dict.get('INCOMPATIBLE_MODELS', '') 
+
+        preferred_list = _string_to_list(preferred_str)
+        incompatible_list = _string_to_list(incompatible_str)
+
+        models = found_models.get("resources", [])
+        if not models:
+            logger.error("No models found.")
+        else:
+            # Remove incompatible models
+            filtered_models = []
+            for model in models:
+                model_id = model.get("model_id", "")
+                short_desc = model.get("short_description", "")
+                if any(incomp in model_id.lower() for incomp in incompatible_list):
+                    continue
+                if any(incomp in short_desc.lower() for incomp in incompatible_list):
+                    continue
+                filtered_models.append(model)
+            
+            # Sort to put the preferred first
+            def sort_key(model):
+                model_id = model.get("model_id", "").lower()
+                is_preferred = any(pref in model_id for pref in preferred_list)
+                return (0 if is_preferred else 1, model_id)
+            
+            sorted_models = sorted(filtered_models, key=sort_key)
+            
+            if print_raw:
+                theme = rich.theme.Theme({"model.name": "bold cyan"})
+                console = rich.console.Console(highlighter=ModelHighlighter(), theme=theme)
+                console.print("[bold]Available Models:[/bold]")
+
+                for model in (virtual_models + virtual_model_policies):
+                    console.print(f"- ✨️ {model.name}:", model.description or 'No description provided.')
+
+                for model in sorted_models:
+                    model_id = model.get("model_id", "N/A")
+                    short_desc = model.get("short_description", "No description provided.")
+                    full_model_name = f"watsonx/{model_id}: {short_desc}"
+                    marker = "★ " if any(pref in model_id.lower() for pref in preferred_list) else ""
+                    console.print(f"- [yellow]{marker}[/yellow]{full_model_name}")
+
+                console.print("[yellow]★[/yellow] [italic dim]indicates a supported and preferred model[/italic dim]\n[blue dim]✨️[/blue dim] [italic dim]indicates a model from a custom provider[/italic dim]" )
+            else:
+                table = rich.table.Table(
+                    show_header=True,
+                    title="[bold]Available Models[/bold]",
+                    caption="[yellow]★ [/yellow] indicates a supported and preferred model from watsonx\n[blue]✨️[/blue] indicates a model from a custom provider",
+                    show_lines=True)
+                columns = ["Model", "Description"]
+                for col in columns:
+                    table.add_column(col)
+
+                for model in (virtual_models + virtual_model_policies):
+                    table.add_row(f"✨️ {model.name}", model.description or 'No description provided.')
+
+                for model in sorted_models:
+                    model_id = model.get("model_id", "N/A")
+                    short_desc = model.get("short_description", "No description provided.")
+                    marker = "★ " if any(pref in model_id.lower() for pref in preferred_list) else ""
+                    table.add_row(f"[yellow]{marker}[/yellow]watsonx/{model_id}", short_desc)
+            
+                rich.print(table)
+
+    def import_model(self, file: str, app_id: str | None) -> List[VirtualModel]:
+        from ibm_watsonx_orchestrate.cli.commands.models.model_provider_mapper import validate_ProviderConfig # lazily import this because the lut building is expensive
+        models = parse_model_file(file)
+
+        for model in models:
+            if not model.name.startswith('virtual-model/'):
+                model.name = f"virtual-model/{model.name}"
+            
+            provider = next(filter(lambda x: x not in ('virtual-policy', 'virtual-model'), model.name.split('/')))
+            if not model.provider_config:   
+                model.provider_config = ProviderConfig.model_validate({"provider": provider})
+            else:
+                model.provider_config.provider = provider
+
+            if "anthropic" in model.name:
+                if not model.config:
+                    model.config = {}
+                if "max_tokens" not in model.config:
+                    model.config["max_tokens"] = ANTHROPIC_DEFAULT_MAX_TOKENS
+
+            if app_id:
+                model.connection_id = get_connection_id(app_id, supported_schemas={ConnectionType.KEY_VALUE})
+            validate_ProviderConfig(model.provider_config, app_id=app_id)
+        return models
+
+    def create_model(self, name: str, display_name: str | None = None, description: str | None = None, provider_config_dict: dict = None, model_type: ModelType = ModelType.CHAT, app_id: str = None) -> VirtualModel:
+        from ibm_watsonx_orchestrate.cli.commands.models.model_provider_mapper import validate_ProviderConfig # lazily import this because the lut building is expensive
+        
+        provider =next(filter(lambda x: x not in ('virtual-policy', 'virtual-model'), name.split('/')))
+
+        provider_config = {}
+        if provider_config_dict:
+            provider_config = ProviderConfig.model_validate(provider_config_dict)
+            provider_config.provider = provider
+        else:
+            provider_config = ProviderConfig.model_validate({"provider": provider})
+        validate_ProviderConfig(provider_config, app_id=app_id)
+
+        if not name.startswith('virtual-model/'):
+            name = f"virtual-model/{name}"
+        
+        config=None
+        # Anthropic has no default for max_tokens
+        if "anthropic" in name:
+            config = {
+                "max_tokens": ANTHROPIC_DEFAULT_MAX_TOKENS
+            }
+
+        model = VirtualModel(
+            name=name,
+            display_name=display_name,
+            description=description,
+            tags=[],
+            provider_config=provider_config,
+            config=config,
+            model_type=model_type,
+            connection_id=get_connection_id(app_id, supported_schemas={ConnectionType.KEY_VALUE})
+        )
+
+        return model
+
+    def publish_or_update_models(self, model: VirtualModel) -> None:
+        models_client = self.get_models_client()
+
+        existing_models = models_client.get_draft_by_name(model.name)
+        if len(existing_models) > 1:
+            logger.error(f"Multiple models with the name '{model.name}' found. Failed to update model")
+            sys.exit(1)
+
+        if len(existing_models) == 1:
+            self.update_model(model_id=existing_models[0].id, model=model)
+        else:
+            self.publish_model(model=model)
+    
+    def publish_model(self, model: VirtualModel) -> None:
+        self.get_models_client().create(model)
+        logger.info(f"Successfully added the model '{model.name}'")
+
+    def update_model(self, model_id: str, model: VirtualModel) -> None:
+        logger.info(f"Existing model '{model.name}' found. Updating...")
+        self.get_models_client().update(model_id, model)
+        logger.info(f"Model '{model.name}' updated successfully")
+    
+    def remove_model(self, name: str) -> None:
+        models_client: ModelsClient = self.get_models_client()
+       
+        existing_models = models_client.get_draft_by_name(name)
+
+        if len(existing_models) > 1:
+            logger.error(f"Multiple models with the name '{name}' found. Failed to remove model")
+            sys.exit(1)
+        if len(existing_models) == 0:
+            logger.error(f"No model found with the name '{name}'")
+            sys.exit(1)
+        
+        model = existing_models[0]
+
+        models_client.delete(model_id=model.id)
+        logger.info(f"Successfully removed the model '{name}'")
+
+    def import_model_policy(self, file: str) -> List[ModelPolicy]:
+        policies = parse_policy_file(file)
+        model_client: ModelsClient = self.get_models_client()
+        model_lut = {m.name: m.id for m in model_client.list()}
+
+        for policy in policies:
+            models =  get_model_names_from_policy(policy)
+            for m in models:
+                if m not in model_lut:
+                    logger.error(f"No model found with the name '{m}'")
+                    sys.exit(1)
+        
+            if not policy.name.startswith('virtual-policy/'):
+                policy.name = f"virtual-policy/{policy.name}"
+
+        return policies
+
+    def create_model_policy(
+        self,
+        name: str,
+        models: List[str],
+        strategy: ModelPolicyStrategyMode, 
+        strategy_on_code: List[int],
+        retry_on_code: List[int],
+        retry_attempts: int,
+        display_name: str = None,
+        description: str = None
+    ) -> ModelPolicy:
+        
+        model_client: ModelsClient = self.get_models_client()
+        model_lut = {m.name: m.id for m in model_client.list()}
+        for m in models:
+            if m not in model_lut:
+                logger.error(f"No model found with the name '{m}'")
+                sys.exit(1)
+        
+        if not name.startswith('virtual-policy/'):
+            name = f"virtual-policy/{name}"
+
+        inner = ModelPolicyInner()
+        inner.strategy = ModelPolicyStrategy(
+            mode=strategy,
+            on_status_codes=strategy_on_code
+        )
+        inner.targets = [ModelPolicyTarget(weight=1, model_name=m) for m in models]
+        if retry_on_code:
+            inner.retry = ModelPolicyRetry(
+                on_status_codes=retry_on_code,
+                attempts=retry_attempts
+            )
+
+        policy = ModelPolicy(
+            name=name,
+            display_name=display_name or name,
+            description=description or name,
+            policy=inner
+        )
+
+        return policy
+
+    def publish_or_update_model_policies(self, policy: ModelPolicy) -> None:
+        model_policies_client: ModelPoliciesClient = self.get_model_policies_client()
+
+        existing_policies = model_policies_client.get_draft_by_name(policy.name)
+        if len(existing_policies) > 1:
+            logger.error(f"Multiple model policies with the name '{policy.name}' found. Failed to update model policy")
+            sys.exit(1)
+
+        if len(existing_policies) == 1:
+            self.update_policy(policy_id=existing_policies[0].id, policy=policy)
+        else:
+            self.publish_policy(policy=policy)
+    
+    def publish_policy(self, policy: VirtualModel) -> None:
+        self.get_model_policies_client().create(policy)
+        logger.info(f"Successfully added the model policy '{policy.name}'")
+
+    def update_policy(self, policy_id: str, policy: VirtualModel) -> None:
+        logger.info(f"Existing model policy '{policy.name}' found. Updating...")
+        self.get_model_policies_client().update(policy_id, policy)
+        logger.info(f"Model policy '{policy.name}' updated successfully")
+    
+    def remove_policy(self, name: str) -> None:
+        model_policies_client: ModelPoliciesClient = self.get_model_policies_client()
+        existing_model_policies = model_policies_client.get_draft_by_name(name)
+
+        if len(existing_model_policies) > 1:
+            logger.error(f"Multiple model policies with the name '{name}' found. Failed to remove model policy")
+            sys.exit(1)
+        if len(existing_model_policies) == 0:
+            logger.error(f"No model policy found with the name '{name}'")
+            sys.exit(1)
+
+        policy = existing_model_policies[0]
+
+        model_policies_client.delete(model_policy_id=policy.id)
+        logger.info(f"Successfully removed the policy '{name}'")

ibm_watsonx_orchestrate/cli/commands/server/server_command.py

@@ -22,6 +22,7 @@ from ibm_watsonx_orchestrate.cli.commands.environment.environment_controller imp
 
 from ibm_watsonx_orchestrate.cli.config import LICENSE_HEADER, \
     ENV_ACCEPT_LICENSE
+
 from ibm_watsonx_orchestrate.cli.config import PROTECTED_ENV_NAME, clear_protected_env_credentials_token, Config, \
     AUTH_CONFIG_FILE_FOLDER, AUTH_CONFIG_FILE, AUTH_MCSP_TOKEN_OPT, AUTH_SECTION_HEADER, USER_ENV_CACHE_HEADER, LICENSE_HEADER, \
     ENV_ACCEPT_LICENSE
@@ -816,4 +817,4 @@ def run_db_migration() -> None:
         sys.exit(1)
 
 if __name__ == "__main__":
-    server_app()
+    server_app()

ibm_watsonx_orchestrate/cli/commands/tools/tools_controller.py

@@ -25,7 +25,7 @@ from rich.panel import Panel
 
 from ibm_watsonx_orchestrate.agent_builder.tools import BaseTool, ToolSpec
 from ibm_watsonx_orchestrate.agent_builder.tools.openapi_tool import create_openapi_json_tools_from_uri,create_openapi_json_tools_from_content
-from ibm_watsonx_orchestrate.cli.commands.models.models_command import ModelHighlighter
+from ibm_watsonx_orchestrate.cli.commands.models.models_controller import ModelHighlighter
 from ibm_watsonx_orchestrate.cli.commands.tools.types import RegistryType
 from ibm_watsonx_orchestrate.cli.commands.connections.connections_controller import configure_connection, remove_connection, add_connection
 from ibm_watsonx_orchestrate.agent_builder.connections.types import  ConnectionType, ConnectionEnvironment, ConnectionPreference

ibm_watsonx_orchestrate/client/connections/__init__.py

@@ -6,5 +6,6 @@ from .connections_client import (
 
 from .utils import (
     get_connections_client,
-    get_connection_type
+    get_connection_type,
+    get_connection_id
 )

ibm_watsonx_orchestrate/client/connections/utils.py

@@ -1,8 +1,11 @@
+import logging
 from ibm_watsonx_orchestrate.client.utils import instantiate_client, is_local_dev
 from ibm_watsonx_orchestrate.client.connections.connections_client import ConnectionsClient
 from ibm_watsonx_orchestrate.cli.config import Config, ENVIRONMENTS_SECTION_HEADER, CONTEXT_SECTION_HEADER, CONTEXT_ACTIVE_ENV_OPT, ENV_WXO_URL_OPT
 from ibm_watsonx_orchestrate.agent_builder.connections.types import ConnectionType, ConnectionAuthType, ConnectionSecurityScheme
 
+logger = logging.getLogger(__name__)
+
 LOCAL_CONNECTION_MANAGER_PORT = 3001
 
 def _get_connections_manager_url() -> str:
@@ -25,3 +28,30 @@ def get_connection_type(security_scheme: ConnectionSecurityScheme, auth_type: Co
     if security_scheme != ConnectionSecurityScheme.OAUTH2:
         return ConnectionType(security_scheme)
     return ConnectionType(auth_type)
+
+def get_connection_id(app_id: str, supported_schemas: set | None = None) -> str:
+    if not app_id:
+        return
+    
+    connections_client = get_connections_client()
+
+    connection_id = None
+    if app_id is not None:
+        connection = connections_client.get(app_id=app_id)
+        if  not connection:
+            logger.error(f"No connection exists with the app-id '{app_id}'")
+            exit(1)
+        connection_id = connection.connection_id
+    
+    existing_draft_configuration = None
+    existing_live_configuration = None
+
+    if supported_schemas:
+        existing_draft_configuration = connections_client.get_config(app_id=app_id, env='draft')
+        existing_live_configuration = connections_client.get_config(app_id=app_id, env='live')
+        for config in [existing_draft_configuration, existing_live_configuration]:
+            if config and config.security_scheme not in supported_schemas:
+                logger.error(f"Only {', '.join(supported_schemas)} credentials are currently supported. Provided connection '{config.app_id}' is of type '{config.security_scheme}' in environment '{config.environment}'")
+                exit(1)
+
+    return connection_id

ibm_watsonx_orchestrate/client/model_policies/model_policies_client.py

@@ -5,15 +5,12 @@ from ibm_watsonx_orchestrate.client.base_api_client import BaseAPIClient, Client
 
 import logging
 
-from ibm_watsonx_orchestrate.client.model_policies.types import ModelPolicy
-from ibm_watsonx_orchestrate.client.models.types import ListVirtualModel, CreateVirtualModel
+from ibm_watsonx_orchestrate.agent_builder.model_policies.types import ModelPolicy
 
 logger = logging.getLogger(__name__)
 
 
 
-
-
 class ModelPoliciesClient(BaseAPIClient):
     """
     Client to handle CRUD operations for ModelPolicies endpoint
@@ -21,6 +18,10 @@ class ModelPoliciesClient(BaseAPIClient):
     # POST api/v1/model_policy
     def create(self, model: ModelPolicy) -> None:
         self._post("/model_policy", data=model.model_dump())
+    
+    # PUT api/v1/model_policy/{model_policy_id}
+    def update(self, model_policy_id: str, model: ModelPolicy) -> None:
+        self._put(f"/model_policy/{model_policy_id}", data=model.model_dump())
 
     # DELETE api/v1/model_policy/{model_policy_id}
     def delete(self, model_policy_id: str) -> dict:
@@ -43,5 +44,23 @@ class ModelPoliciesClient(BaseAPIClient):
             if e.response.status_code == 404:
                 return []
             raise e
+    
+    def get_drafts_by_names(self, policy_names: List[str]) -> List[ModelPolicy]:
+        try:
+            formatted_policy_names = [f'name={x}' for x  in policy_names]
+            res = self._get(f"/model_policy?{'&'.join(formatted_policy_names)}")
+            return [ModelPolicy.model_validate(conn) for conn in res]
+        except ValidationError as e:
+            logger.error("Received unexpected response from server")
+            raise e
+        except ClientAPIException as e:
+            if e.response.status_code == 404:
+                return []
+            raise e
+    
+    def get_draft_by_name(self, policy_name: str) -> ModelPolicy:
+        return self.get_drafts_by_names([policy_name])
+
+
 
 

ibm_watsonx_orchestrate/client/models/models_client.py

@@ -5,7 +5,7 @@ from ibm_watsonx_orchestrate.client.base_api_client import BaseAPIClient, Client
 
 import logging
 
-from ibm_watsonx_orchestrate.client.models.types import ListVirtualModel, CreateVirtualModel
+from ibm_watsonx_orchestrate.agent_builder.models.types import ListVirtualModel, VirtualModel
 
 logger = logging.getLogger(__name__)
 
@@ -18,8 +18,12 @@ class ModelsClient(BaseAPIClient):
     Client to handle CRUD operations for Models endpoint
     """
     # POST api/v1/models
-    def create(self, model: CreateVirtualModel) -> None:
-        self._post("/models", data=model.model_dump())
+    def create(self, model: VirtualModel) -> None:
+        self._post("/models", data=model.model_dump(exclude_none=True))
+    
+    # PUT api/v1/models/{models_id}
+    def update(self, model_id: str, model: VirtualModel) -> None:
+        self._put(f"/models/{model_id}", data=model.model_dump(exclude_none=True))
 
     # DELETE api/v1/models/{model_id}
     def delete(self, model_id: str) -> dict:
@@ -42,5 +46,21 @@ class ModelsClient(BaseAPIClient):
             if e.response.status_code == 404:
                 return []
             raise e
+    
+    def get_drafts_by_names(self, model_names: List[str]) -> List[ListVirtualModel]:
+        try:
+            formatted_model_names = [f'name={x}' for x  in model_names]
+            res = self._get(f"/models?{'&'.join(formatted_model_names)}")
+            return [ListVirtualModel.model_validate(conn) for conn in res]
+        except ValidationError as e:
+            logger.error("Received unexpected response from server")
+            raise e
+        except ClientAPIException as e:
+            if e.response.status_code == 404:
+                return []
+            raise e
+    
+    def get_draft_by_name(self, model_name: str) -> ListVirtualModel:
+        return self.get_drafts_by_names([model_name])
 
 

ibm_watsonx_orchestrate/client/tools/tool_client.py

@@ -1,3 +1,4 @@
+from typing import Literal
 from ibm_watsonx_orchestrate.client.base_api_client import BaseAPIClient, ClientAPIException
 from typing_extensions import List
 
@@ -32,7 +33,7 @@ class ToolClient(BaseAPIClient):
         formatted_tool_names = [f"names={x}" for x in tool_names]
         return self._get(f"/tools?{'&'.join(formatted_tool_names)}")
     
-    def get_draft_by_id(self, tool_id: str) -> List[dict]:
+    def get_draft_by_id(self, tool_id: str) -> dict | Literal[""]:
         if tool_id is None:
             return ""
         else:

ibm_watsonx_orchestrate/docker/compose-lite.yml

@@ -323,6 +323,7 @@ services:
       AI_GATEWAY_BASE_URL: ${AI_GATEWAY_BASE_URL}
       AI_GATEWAY_ENABLED : ${AI_GATEWAY_ENABLED}
 
+
   wxo-server-worker:
     image: ${WORKER_REGISTRY:-us.icr.io/watson-orchestrate-private}/wxo-server-conversation_controller:${WORKER_TAG:-latest}
     platform: linux/amd64
@@ -450,6 +451,7 @@ services:
       STORAGE_S3_REGION: us-east-1
       AWS_ACCESS_KEY_ID: ${MINIO_ROOT_USER:-minioadmin}
       AWS_SECRET_ACCESS_KEY: ${MINIO_ROOT_PASSWORD:-watsonxorchestrate}
+      TEMPUS_HOST_NAME: http://wxo-tempus-runtime:9044
     extra_hosts:
       - "host.docker.internal:host-gateway"