Skip to content

Add conditional chat incrmenet step #7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
adi-kmt wants to merge 1 commit into
base: main
Choose a base branch
from
Draft

Add conditional chat incrmenet step #7

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
36 changes: 31 additions & 5 deletions minference/lite/requests.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -218,6 +218,16 @@ def validate_vllm_request(request: Dict[str, Any]) -> bool:
EntityRegistry._logger.error(f"Error validating VLLM request: {e}")
raise ValueError(f"Error validating VLLM request: {e} with request: {request}")

def reset_workflow_step(chat_thread: ChatThread, step: int = 0):
"""Resets the workflow step for a chat thread.

Args:
chat_thread: The ChatThread object.
step: The step to reset to (default: 0).
"""
chat_thread.workflow_step = step
EntityRegistry._logger.info(f"Reset workflow step for ChatThread({chat_thread.id}) to {step}")

def get_openai_request(chat_thread: ChatThread) -> Optional[Dict[str, Any]]:
"""Get OpenAI format request from chat thread."""
EntityRegistry._logger.info(f"Getting OpenAI request for ChatThread({chat_thread.id}) with response format {chat_thread.llm_config.response_format}")
Expand All @@ -242,7 +252,7 @@ def get_openai_request(chat_thread: ChatThread) -> Optional[Dict[str, Any]]:
request["tool_choice"] = "auto"
EntityRegistry._logger.info(f"Added {len(tools)} tools to OpenAI request as auto_tools")
elif chat_thread.llm_config.response_format == ResponseFormat.workflow:
#detected workflow mode
#detected workflow mode
if chat_thread.workflow_step is None:
raise ValueError("Workflow step is None")
EntityRegistry._logger.info(f"Detected workflow mode for ChatThread({chat_thread.id}) with workflow step {chat_thread.workflow_step}")
Expand All @@ -253,12 +263,22 @@ def get_openai_request(chat_thread: ChatThread) -> Optional[Dict[str, Any]]:
request["tools"] = [tool.get_openai_tool()]
request["tool_choice"] = {"type": "function", "function": {"name": tool.name}}
EntityRegistry._logger.info(f"Added tool({tool.name}) to OpenAI request as workflow step {chat_thread.workflow_step}")
chat_thread.workflow_step += 1
# Increment ONLY if validation succeeds
if validate_openai_request(request):
chat_thread.workflow_step += 1
EntityRegistry._logger.info(f"Validated OpenAI request for ChatThread({chat_thread.id}) with response format {chat_thread.llm_config.response_format}")
return request
else:
EntityRegistry._logger.error(f"Failed to validate OpenAI request for ChatThread({chat_thread.id}) with response format {chat_thread.llm_config.response_format}")
return None

else:
EntityRegistry._logger.error(f"Tool not found for workflow step {chat_thread.workflow_step}")
return None # Return None if tool not found

elif chat_thread.llm_config.response_format != ResponseFormat.text:
raise ValueError(f"Invalid response format: {chat_thread.llm_config.response_format}")

if validate_openai_request(request):
EntityRegistry._logger.info(f"Validated OpenAI request for ChatThread({chat_thread.id}) with response format {chat_thread.llm_config.response_format}")
return request
Expand All @@ -268,7 +288,7 @@ def get_openai_request(chat_thread: ChatThread) -> Optional[Dict[str, Any]]:

def get_anthropic_request(chat_thread: ChatThread) -> Optional[Dict[str, Any]]:
"""Get Anthropic format request from chat thread."""
system_content, messages = chat_thread.anthropic_messages
system_content, messages = chat_thread.anthropic_messages
request = {
"model": chat_thread.llm_config.model,
"max_tokens": chat_thread.llm_config.max_tokens,
Expand All @@ -295,9 +315,15 @@ def get_anthropic_request(chat_thread: ChatThread) -> Optional[Dict[str, Any]]:
if tool:
request["tools"] = [tool.get_anthropic_tool()]
request["tool_choice"] = ToolChoiceToolChoiceTool(name=tool.name, type="tool")
chat_thread.workflow_step += 1
# Increment ONLY if validation succeeds
if validate_anthropic_request(request):
chat_thread.workflow_step += 1
return request
else:
return None
else:
EntityRegistry._logger.error(f"Tool not found for workflow step {chat_thread.workflow_step}")
return None # Return None if tool not found



Expand Down