versionhq 1.1.12.5__py3-none-any.whl → 1.1.13.1__py3-none-any.whl

versionhq/__init__.py

@@ -4,16 +4,22 @@ warnings.filterwarnings(action="ignore", message="Pydantic serializer warnings:"
 warnings.filterwarnings(action="ignore", category=UserWarning, module="pydantic._internal")
 warnings.filterwarnings(action="ignore", module="LiteLLM:utils")
 
+from dotenv import load_dotenv
+load_dotenv(override=True)
+
 from versionhq.agent.model import Agent
+from versionhq.llm.model import LLM
+from versionhq.llm.llm_vars import LLM_CONTEXT_WINDOW_SIZES, PARAMS, PROVIDERS, MODELS
 from versionhq.clients.customer.model import Customer
 from versionhq.clients.product.model import Product, ProductProvider
 from versionhq.clients.workflow.model import MessagingWorkflow, MessagingComponent
 from versionhq.knowledge.model import Knowledge, KnowledgeStorage
 from versionhq.knowledge.source import PDFKnowledgeSource, CSVKnowledgeSource, JSONKnowledgeSource, TextFileKnowledgeSource, ExcelKnowledgeSource, StringKnowledgeSource
 from versionhq.knowledge.source_docling import DoclingSource
+from versionhq.network.model import TaskStatus, TaskGraph, Node, Edge, DependencyType
 from versionhq.task.model import Task, TaskOutput, ConditionalTask, ResponseField
 from versionhq.task.evaluate import Evaluation, EvaluationItem
-from versionhq.team.model import Team, TeamOutput, Formation, TeamMember, TaskHandlingProcess
+from versionhq.team.model import Team, TeamOutput, Formation, Member, TaskHandlingProcess
 from versionhq.tool.model import Tool, ToolSet
 from versionhq.tool.cache_handler import CacheHandler
 from versionhq.tool.tool_handler import ToolHandler
@@ -24,10 +30,16 @@ from versionhq.memory.model import ShortTermMemory,LongTermMemory, UserMemory, M
 from versionhq.task.formation import form_agent_network
 
 
-__version__ = "1.1.12.5"
+__version__ = "1.1.13.1"
 __all__ = [
     "Agent",
 
+    "LLM",
+    "LLM_CONTEXT_WINDOW_SIZES",
+    "PARAMS",
+    "PROVIDERS",
+    "MODELS",
+
     "Customer",
     "Product",
     "ProductProvider",
@@ -44,6 +56,12 @@ __all__ = [
     "StringKnowledgeSource",
     "DoclingSource",
 
+    "TaskStatus",
+    "TaskGraph",
+    "Node",
+    "Edge",
+    "DependencyType",
+
     "Task",
     "TaskOutput",
     "ConditionalTask",
@@ -55,7 +73,7 @@ __all__ = [
     "Team",
     "TeamOutput",
     "Formation",
-    "TeamMember",
+    "Member",
     "TaskHandlingProcess",
 
     "Tool",

versionhq/_utils/usage_metrics.py

@@ -3,14 +3,14 @@ from pydantic import BaseModel, Field
 
 class UsageMetrics(BaseModel):
     """
-    Model to track usage metrics for the agent/team's execution.
+    Model to track usage
     """
 
-    total_tokens: int = Field(default=0, description="Total number of tokens used")
-    prompt_tokens: int = Field(default=0, description="Number of tokens used in prompts")
-    cached_prompt_tokens: int = Field(default=0, description="Number of cached prompt tokens used")
-    completion_tokens: int = Field(default=0, description="Number of tokens used in completions")
-    successful_requests: int = Field(default=0, description="Number of successful requests made")
+    total_tokens: int = Field(default=0, description="total number of tokens used")
+    prompt_tokens: int = Field(default=0, description="number of tokens used in prompts")
+    cached_prompt_tokens: int = Field(default=0, description="number of cached prompt tokens used")
+    completion_tokens: int = Field(default=0, description="number of tokens used in completions")
+    successful_requests: int = Field(default=0, description="number of successful requests made")
 
     def add_usage_metrics(self, usage_metrics: "UsageMetrics") -> None:
         """

versionhq/agent/inhouse_agents.py

@@ -29,7 +29,7 @@ vhq_formation_planner = Agent(
     role="vhq-Formation Planner",
     goal="Plan a formation of agents based on the given task descirption.",
     llm="gemini/gemini-2.0-flash-exp",
-    llm_config=dict(top_p=0.8, top_k=30, temperature=0.9),
+    llm_config=dict(top_p=0.8, topK=40, temperature=0.9),
     maxit=1,
     max_retry_limit=1,
     knowledge_sources=[

versionhq/agent/model.py

@@ -23,25 +23,6 @@ load_dotenv(override=True)
 T = TypeVar("T", bound="Agent")
 
 
-# def mock_agent_ops_provider():
-#     def track_agent(*args, **kwargs):
-#         def noop(f):
-#             return f
-#         return noop
-#     return track_agent
-
-# track_agent = mock_agent_ops_provider()
-
-# agentops = None
-# if os.environ.get("AGENTOPS_API_KEY"):
-#     try:
-#         from agentops import track_agent
-#     except ImportError:
-#         track_agent = mock_agent_ops_provider()
-# else:
-#    track_agent = mock_agent_ops_provider()
-
-
 class TokenProcess:
     total_tokens: int = 0
     prompt_tokens: int = 0
@@ -76,10 +57,9 @@ class TokenProcess:
 # @track_agent()
 class Agent(BaseModel):
     """
-    Agent class that run on LLM.
-    Agents execute tasks alone or in the team, using RAG tools and knowledge base if any.
-    Agents will prioritize team tasks when they belong to the team.
-    * (Temp) Comment out all the optional fields except for Team and LLM settings for convenience.
+    A class to store agent information.
+    Agents must have `role`, `goal`, and `llm` = DEFAULT_MODEL_NAME as default.
+    Then run validation on `backstory`, `llm`, `tools`, `rpm` (request per min), `knowledge`, and `memory`.
     """
 
     __hash__ = object.__hash__
@@ -132,8 +112,6 @@ class Agent(BaseModel):
 
     # cache, error, ops handling
     formatting_errors: int = Field(default=0, description="number of formatting errors.")
-    agent_ops_agent_name: str = None
-    agent_ops_agent_id: str = None
 
 
     @field_validator("id", mode="before")
@@ -161,9 +139,8 @@ class Agent(BaseModel):
     @model_validator(mode="after")
     def set_up_llm(self) -> Self:
         """
-        Set up `llm` and `function_calling_llm` as valid LLM objects using the given values.
+        Set up `llm` and `function_calling_llm` as valid LLM objects using the given kwargs.
         """
-        self.agent_ops_agent_name = self.role
         self.llm = self._convert_to_llm_object(llm=self.llm)
 
         function_calling_llm = self.function_calling_llm if self.function_calling_llm else self.llm if self.llm else None
@@ -177,12 +154,15 @@ class Agent(BaseModel):
         return self
 
 
-    def _convert_to_llm_object(self, llm: Any | None) -> LLM:
+    def _convert_to_llm_object(self, llm: Any = None) -> LLM:
         """
         Convert the given value to LLM object.
         When `llm` is dict or self.llm_config is not None, add these values to the LLM object after validating them.
         """
-        llm = llm if llm is not None else DEFAULT_MODEL_NAME
+        llm = llm if llm else self.llm if self.llm else DEFAULT_MODEL_NAME
+
+        if not llm:
+            pass
 
         match llm:
             case LLM():
@@ -380,16 +360,16 @@ class Agent(BaseModel):
         Set up memories: stm, ltm, and um
         """
 
-        if self.use_memory == True:
-            self.long_term_memory = self.long_term_memory if self.long_term_memory else LongTermMemory()
-            self.short_term_memory = self.short_term_memory if self.short_term_memory else ShortTermMemory(agent=self, embedder_config=self.embedder_config)
+        # if self.use_memory == True:
+        self.long_term_memory = self.long_term_memory if self.long_term_memory else LongTermMemory()
+        self.short_term_memory = self.short_term_memory if self.short_term_memory else ShortTermMemory(agent=self, embedder_config=self.embedder_config)
 
-            if hasattr(self, "memory_config") and self.memory_config is not None:
-                user_id = self.memory_config.get("user_id", None)
-                if user_id:
-                    self.user_memory = self.user_memory if self.user_memory else UserMemory(agent=self, user_id=user_id)
-            else:
-                self.user_memory = None
+        if hasattr(self, "memory_config") and self.memory_config is not None:
+            user_id = self.memory_config.get("user_id", None)
+            if user_id:
+                self.user_memory = self.user_memory if self.user_memory else UserMemory(agent=self, user_id=user_id)
+        else:
+            self.user_memory = None
 
         return self
 
@@ -402,6 +382,78 @@ class Agent(BaseModel):
             pass
 
 
+    def update_llm(self, llm: Any = None, llm_config: Optional[Dict[str, Any]] = None) -> Self:
+        """
+        Update llm and llm_config of the exsiting agent. (Other conditions will remain the same.)
+        """
+
+        if not llm and not llm_config:
+            self._logger.log(level="error", message="Missing llm or llm_config values to update", color="red")
+            pass
+
+        self.llm = llm
+        if llm_config:
+            if self.llm_config:
+                self.llm_config.update(llm_config)
+            else:
+                self.llm_config = llm_config
+
+        return self.set_up_llm()
+
+
+    def update(self, **kwargs) -> Self:
+        """
+        Update the existing agent. Address variables that require runnning set_up_x methods first, then update remaining variables.
+        """
+
+        if not kwargs:
+            self._logger.log(level="error", message="Missing values to update", color="red")
+            return self
+
+        for k, v in kwargs.items():
+            match k:
+                case "tools":
+                    self.tools = kwargs.get(k, self.tools)
+                    self.set_up_tools()
+
+                case "role" | "goal":
+                    self.role = kwargs.get("role", self.role)
+                    self.goal = kwargs.get("goal", self.goal)
+                    if not self.backstory:
+                        self.set_up_backstory()
+
+                    if self.backstory:
+                        self.backstory += f"new role: {self.role}, new goal: {self.goal}"
+
+                case "max_rpm":
+                    self.max_rpm = kwargs.get(k, self.max_rpm)
+                    self.set_up_rpm()
+
+                case "knowledge_sources":
+                    self.knowledge_sources = kwargs.get("knowledge_sources", self.knowledge_sources)
+                    self.set_up_knowledge()
+
+                case "use_memory" | "memory_config":
+                    self.use_memory = kwargs.get("use_memory", self.use_memory)
+                    self.memory_config = kwargs.get("memory_config", self.memory_config)
+                    self.set_up_memory()
+
+                case "llm" | "llm_config":
+                    self.llm = kwargs.get("llm", self.llm)
+                    self.llm_config = kwargs.get("llm_config", self.llm_config)
+                    self.update_llm(llm=self.llm, llm_config=self.llm_config)
+
+                case _:
+                    try:
+                        setattr(self, k, v)
+                    except Exception as e:
+                        self._logger.log(level="error", message=f"Failed to update the key: {k} We'll skip. Error: {str(e)}", color="red")
+                        pass
+
+        return self
+
+
+
     def invoke(
         self,
         prompts: str,

versionhq/llm/llm_vars.py

@@ -1,4 +1,3 @@
-from enum import Enum
 from typing import Type
 
 JSON_URL = "https://raw.githubusercontent.com/BerriAI/litellm/main/model_prices_and_context_window.json"
@@ -55,6 +54,11 @@ MODELS = {
     "openrouter": [
         "openrouter/deepseek/deepseek-r1",
         "openrouter/qwen/qwen-2.5-72b-instruct",
+        "openrouter/google/gemini-2.0-flash-thinking-exp:free",
+        "openrouter/google/gemini-2.0-flash-thinking-exp-1219:free",
+        "openrouter/google/gemini-2.0-flash-001",
+        "openrouter/meta-llama/llama-3.3-70b-instruct",
+        "openrouter/mistralai/mistral-large-2411",
     ],
     "huggingface": [
         "huggingface/qwen/qwen2.5-VL-72B-Instruct",

versionhq/llm/model.py

@@ -111,7 +111,7 @@ class LLM(BaseModel):
 
 
     @model_validator(mode="after")
-    def validate_model_providers(self) -> Self:
+    def validate_model_and_provider(self) -> Self:
         """
         Validate the given model, provider, interface provider.
         """