github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-06-23 15:19:33 +00:00
Code Issues Packages Projects Releases Wiki Activity

Move OAI assistants to langchain and add callbacks (#13236)

Browse Source
This commit is contained in:
Bagatur 2023-11-13 17:42:07 -08:00 committed by GitHub
parent 8006919e52
commit 1c67db4c18
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
5 changed files with 611 additions and 350 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
cookbook/openai_v1_cookbook.ipynb
View File

@ -17,7 +17,7 @@
"metadata": {},
"outputs": [],
"source": [
"# need openai>=1.1.0, langchain>=0.0.333, langchain-experimental>=0.0.39\n",
"# need openai>=1.1.0, langchain>=0.0.335, langchain-experimental>=0.0.39\n",
"!pip install -U openai langchain langchain-experimental"
]
},
@ -109,7 +109,7 @@
"metadata": {},
"outputs": [],
"source": [
"from langchain_experimental.openai_assistant import OpenAIAssistantRunnable"
"from langchain.agents.openai_assistant import OpenAIAssistantRunnable"
]
},
{

224
docs/docs/modules/agents/agent_types/openai_assistants.ipynb Normal file
View File

@ -0,0 +1,224 @@
{
"cells": [
{
"cell_type": "markdown",
"id": "ab4ffc65-4ec2-41f5-b225-e8a7a4c3799f",
"metadata": {},
"source": [
"# OpenAI assistants\n",
"\n",
"> The [Assistants API](https://platform.openai.com/docs/assistants/overview) allows you to build AI assistants within your own applications. An Assistant has instructions and can leverage models, tools, and knowledge to respond to user queries. The Assistants API currently supports three types of tools: Code Interpreter, Retrieval, and Function calling\n",
"\n",
"\n",
"You can interact with OpenAI Assistants using OpenAI tools or custom tools. When using exclusively OpenAI tools, you can just invoke the assistant directly and get final answers. When using custom tools, you can run the assistant and tool execution loop using the built-in AgentExecutor or easily write your own executor.\n",
"\n",
"Below we show the different ways to interact with Assistants. As a simple example, let's build a math tutor that can write and run code."
]
},
{
"cell_type": "markdown",
"id": "f2128b7b-01ff-477e-a41c-98026e0180ac",
"metadata": {},
"source": [
"### Using only OpenAI tools"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "aa761a93-caa1-4e56-b901-5ff50a89bc82",
"metadata": {},
"outputs": [],
"source": [
"from langchain.agents.openai_assistant import OpenAIAssistantRunnable"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "5944a18a-95eb-44ce-a66f-5f50db1d3e1f",
"metadata": {},
"outputs": [],
"source": [
"interpreter_assistant = OpenAIAssistantRunnable.create_assistant(\n",
" name=\"langchain assistant\",\n",
" instructions=\"You are a personal math tutor. Write and run code to answer math questions.\",\n",
" tools=[{\"type\": \"code_interpreter\"}],\n",
" model=\"gpt-4-1106-preview\",\n",
")\n",
"output = interpreter_assistant.invoke({\"content\": \"What's 10 - 4 raised to the 2.7\"})\n",
"output"
]
},
{
"cell_type": "markdown",
"id": "2dbaba4e-3f2f-4b5d-bfc9-fc5ad4a05232",
"metadata": {},
"source": [
"### As a LangChain agent with arbitrary tools\n",
"\n",
"Now let's recreate this functionality using our own tools. For this example we'll use the [E2B sandbox runtime tool](https://e2b.dev/docs?ref=landing-page-get-started)."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "b4fb3305-525b-4739-a485-7a686beff5b6",
"metadata": {},
"outputs": [],
"source": [
"!pip install e2b duckduckgo-search"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "cc0cba70-8507-498d-92ac-fe47133db200",
"metadata": {},
"outputs": [],
"source": [
"from langchain.tools import E2BDataAnalysisTool, DuckDuckGoSearchRun\n",
"\n",
"tools = [E2BDataAnalysisTool(api_key=\"...\"), DuckDuckGoSearchRun()]"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "91e6973d-3d9a-477f-99e2-4aaad16004ec",
"metadata": {},
"outputs": [],
"source": [
"agent = OpenAIAssistantRunnable.create_assistant(\n",
" name=\"langchain assistant e2b tool\",\n",
" instructions=\"You are a personal math tutor. Write and run code to answer math questions. You can also search the internet.\",\n",
" tools=tools,\n",
" model=\"gpt-4-1106-preview\",\n",
" as_agent=True,\n",
")"
]
},
{
"cell_type": "markdown",
"id": "78fa9320-06fc-4cbc-a3cf-39aaf2427080",
"metadata": {},
"source": [
"#### Using AgentExecutor"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "e38007a4-fcc1-419b-9ae4-70d36c3fc1cd",
"metadata": {},
"outputs": [],
"source": [
"from langchain.agents import AgentExecutor\n",
"\n",
"agent_executor = AgentExecutor(agent=agent, tools=tools)\n",
"agent_executor.invoke({\"content\": \"What's the weather in SF today divided by 2.7\"})"
]
},
{
"cell_type": "markdown",
"id": "6bf4199a-eed1-485a-8da3-aed948c0e1e2",
"metadata": {},
"source": [
"#### Custom execution"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "357361ff-f54d-4fd0-b69b-77689f56f40e",
"metadata": {},
"outputs": [],
"source": [
"agent = OpenAIAssistantRunnable.create_assistant(\n",
" name=\"langchain assistant e2b tool\",\n",
" instructions=\"You are a personal math tutor. Write and run code to answer math questions.\",\n",
" tools=tools,\n",
" model=\"gpt-4-1106-preview\",\n",
" as_agent=True,\n",
")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "864e7f9b-0501-4bb7-8aad-a7aa19b601af",
"metadata": {},
"outputs": [],
"source": [
"from langchain.schema.agent import AgentFinish\n",
"\n",
"\n",
"def execute_agent(agent, tools, input):\n",
" tool_map = {tool.name: tool for tool in tools}\n",
" response = agent.invoke(input)\n",
" while not isinstance(response, AgentFinish):\n",
" tool_outputs = []\n",
" for action in response:\n",
" tool_output = tool_map[action.tool].invoke(action.tool_input)\n",
" print(action.tool, action.tool_input, tool_output, end=\"\\n\\n\")\n",
" tool_outputs.append(\n",
" {\"output\": tool_output, \"tool_call_id\": action.tool_call_id}\n",
" )\n",
" response = agent.invoke(\n",
" {\n",
" \"tool_outputs\": tool_outputs,\n",
" \"run_id\": action.run_id,\n",
" \"thread_id\": action.thread_id,\n",
" }\n",
" )\n",
"\n",
" return response"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "5ad6bb07-aac4-4b71-9e67-cc177fcbc537",
"metadata": {},
"outputs": [],
"source": [
"response = execute_agent(agent, tools, {\"content\": \"What's 10 - 4 raised to the 2.7\"})\n",
"print(response.return_values[\"output\"])"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "f55a3a3a-8169-491e-aa15-cf30a2b230df",
"metadata": {},
"outputs": [],
"source": [
"next_response = execute_agent(\n",
" agent, tools, {\"content\": \"now add 17.241\", \"thread_id\": response.thread_id}\n",
")\n",
"print(next_response.return_values[\"output\"])"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.9.1"
}
},
"nbformat": 4,
"nbformat_minor": 5
}

354
libs/experimental/langchain_experimental/openai_assistant/base.py
View File

@ -1,350 +1,8 @@
from __future__ import annotations
# flake8: noqa
import json
from time import sleep
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Sequence, Tuple, Union
from langchain.pydantic_v1 import Field
from langchain.schema.agent import AgentAction, AgentFinish
from langchain.schema.runnable import RunnableConfig, RunnableSerializable
from langchain.tools import format_tool_to_openai_function
from langchain.tools.base import BaseTool
if TYPE_CHECKING:
import openai
from openai.types.beta.threads import ThreadMessage
from openai.types.beta.threads.required_action_function_tool_call import (
RequiredActionFunctionToolCall,
)
class OpenAIAssistantFinish(AgentFinish):
"""AgentFinish with run and thread metadata."""
run_id: str
thread_id: str
class OpenAIAssistantAction(AgentAction):
"""AgentAction with info needed to submit custom tool output to existing run."""
tool_call_id: str
run_id: str
thread_id: str
def _get_openai_client() -> openai.OpenAI:
try:
import openai
return openai.OpenAI()
except ImportError as e:
raise ImportError(
"Unable to import openai, please install with `pip install openai`."
) from e
except AttributeError as e:
raise AttributeError(
"Please make sure you are using a v1.1-compatible version of openai. You "
'can install with `pip install "openai>=1.1"`.'
) from e
OutputType = Union[
List[OpenAIAssistantAction],
# For backwards compatibility.
from langchain.agents.openai_assistant.base import (
OpenAIAssistantAction,
OpenAIAssistantFinish,
List["ThreadMessage"],
List["RequiredActionFunctionToolCall"],
]
class OpenAIAssistantRunnable(RunnableSerializable[Dict, OutputType]):
"""Run an OpenAI Assistant.
Example using OpenAI tools:
.. code-block:: python
from langchain_experimental.openai_assistant import OpenAIAssistantRunnable
interpreter_assistant = OpenAIAssistantRunnable.create_assistant(
name="langchain assistant",
instructions="You are a personal math tutor. Write and run code to answer math questions.",
tools=[{"type": "code_interpreter"}],
model="gpt-4-1106-preview"
)
output = interpreter_assistant.invoke({"content": "What's 10 - 4 raised to the 2.7"})
Example using custom tools and AgentExecutor:
.. code-block:: python
from langchain_experimental.openai_assistant import OpenAIAssistantRunnable
from langchain.agents import AgentExecutor
from langchain.tools import E2BDataAnalysisTool
tools = [E2BDataAnalysisTool(api_key="...")]
agent = OpenAIAssistantRunnable.create_assistant(
name="langchain assistant e2b tool",
instructions="You are a personal math tutor. Write and run code to answer math questions.",
tools=tools,
model="gpt-4-1106-preview",
as_agent=True
)
agent_executor = AgentExecutor(agent=agent, tools=tools)
agent_executor.invoke({"content": "What's 10 - 4 raised to the 2.7"})
Example using custom tools and custom execution:
.. code-block:: python
from langchain_experimental.openai_assistant import OpenAIAssistantRunnable
from langchain.agents import AgentExecutor
from langchain.schema.agent import AgentFinish
from langchain.tools import E2BDataAnalysisTool
tools = [E2BDataAnalysisTool(api_key="...")]
agent = OpenAIAssistantRunnable.create_assistant(
name="langchain assistant e2b tool",
instructions="You are a personal math tutor. Write and run code to answer math questions.",
tools=tools,
model="gpt-4-1106-preview",
as_agent=True
)
def execute_agent(agent, tools, input):
tool_map = {tool.name: tool for tool in tools}
response = agent.invoke(input)
while not isinstance(response, AgentFinish):
tool_outputs = []
for action in response:
tool_output = tool_map[action.tool].invoke(action.tool_input)
tool_outputs.append({"output": tool_output, "tool_call_id": action.tool_call_id})
response = agent.invoke(
{
"tool_outputs": tool_outputs,
"run_id": action.run_id,
"thread_id": action.thread_id
}
)
return response
response = execute_agent(agent, tools, {"content": "What's 10 - 4 raised to the 2.7"})
next_response = execute_agent(agent, tools, {"content": "now add 17.241", "thread_id": response.thread_id})
""" # noqa: E501
client: openai.OpenAI = Field(default_factory=_get_openai_client)
"""OpenAI client."""
assistant_id: str
"""OpenAI assistant id."""
check_every_ms: float = 1_000.0
"""Frequency with which to check run progress in ms."""
as_agent: bool = False
"""Use as a LangChain agent, compatible with the AgentExecutor."""
@classmethod
def create_assistant(
cls,
name: str,
instructions: str,
tools: Sequence[Union[BaseTool, dict]],
model: str,
*,
client: Optional[openai.OpenAI] = None,
**kwargs: Any,
) -> OpenAIAssistantRunnable:
"""Create an OpenAI Assistant and instantiate the Runnable.
Args:
name: Assistant name.
instructions: Assistant instructions.
tools: Assistant tools. Can be passed in in OpenAI format or as BaseTools.
model: Assistant model to use.
client: OpenAI client. Will create default client if not specified.
Returns:
OpenAIAssistantRunnable configured to run using the created assistant.
"""
client = client or _get_openai_client()
openai_tools: List = []
for tool in tools:
if isinstance(tool, BaseTool):
tool = {
"type": "function",
"function": format_tool_to_openai_function(tool),
}
openai_tools.append(tool)
assistant = client.beta.assistants.create(
name=name,
instructions=instructions,
tools=openai_tools,
model=model,
)
return cls(assistant_id=assistant.id, **kwargs)
def invoke(
self, input: dict, config: Optional[RunnableConfig] = None
) -> OutputType:
"""Invoke assistant.
Args:
input: Runnable input dict that can have:
content: User message when starting a new run.
thread_id: Existing thread to use.
run_id: Existing run to use. Should only be supplied when providing
the tool output for a required action after an initial invocation.
file_ids: File ids to include in new run. Used for retrieval.
message_metadata: Metadata to associate with new message.
thread_metadata: Metadata to associate with new thread. Only relevant
when new thread being created.
instructions: Additional run instructions.
model: Override Assistant model for this run.
tools: Override Assistant tools for this run.
run_metadata: Metadata to associate with new run.
config: Runnable config:
Return:
If self.as_agent, will return
Union[List[OpenAIAssistantAction], OpenAIAssistantFinish]. Otherwise
will return OpenAI types
Union[List[ThreadMessage], List[RequiredActionFunctionToolCall]].
"""
# Being run within AgentExecutor and there are tool outputs to submit.
if self.as_agent and input.get("intermediate_steps"):
tool_outputs = self._parse_intermediate_steps(input["intermediate_steps"])
run = self.client.beta.threads.runs.submit_tool_outputs(**tool_outputs)
# Starting a new thread and a new run.
elif "thread_id" not in input:
thread = {
"messages": [
{
"role": "user",
"content": input["content"],
"file_ids": input.get("file_ids", []),
"metadata": input.get("message_metadata"),
}
],
"metadata": input.get("thread_metadata"),
}
run = self._create_thread_and_run(input, thread)
# Starting a new run in an existing thread.
elif "run_id" not in input:
_ = self.client.beta.threads.messages.create(
input["thread_id"],
content=input["content"],
role="user",
file_ids=input.get("file_ids", []),
metadata=input.get("message_metadata"),
)
run = self._create_run(input)
# Submitting tool outputs to an existing run, outside the AgentExecutor
# framework.
else:
run = self.client.beta.threads.runs.submit_tool_outputs(**input)
return self._get_response(run.id, run.thread_id)
def _parse_intermediate_steps(
self, intermediate_steps: List[Tuple[OpenAIAssistantAction, str]]
) -> dict:
last_action, last_output = intermediate_steps[-1]
run = self._wait_for_run(last_action.run_id, last_action.thread_id)
required_tool_call_ids = {
tc.id for tc in run.required_action.submit_tool_outputs.tool_calls
}
tool_outputs = [
{"output": output, "tool_call_id": action.tool_call_id}
for action, output in intermediate_steps
if action.tool_call_id in required_tool_call_ids
]
submit_tool_outputs = {
"tool_outputs": tool_outputs,
"run_id": last_action.run_id,
"thread_id": last_action.thread_id,
}
return submit_tool_outputs
def _create_run(self, input: dict) -> Any:
params = {
k: v
for k, v in input.items()
if k in ("instructions", "model", "tools", "run_metadata")
}
return self.client.beta.threads.runs.create(
input["thread_id"],
assistant_id=self.assistant_id,
**params,
)
def _create_thread_and_run(self, input: dict, thread: dict) -> Any:
params = {
k: v
for k, v in input.items()
if k in ("instructions", "model", "tools", "run_metadata")
}
run = self.client.beta.threads.create_and_run(
assistant_id=self.assistant_id,
thread=thread,
**params,
)
return run
def _get_response(self, run_id: str, thread_id: str) -> Any:
# TODO: Pagination
import openai
run = self._wait_for_run(run_id, thread_id)
if run.status == "completed":
messages = self.client.beta.threads.messages.list(thread_id, order="asc")
new_messages = [msg for msg in messages if msg.run_id == run_id]
if not self.as_agent:
return new_messages
answer: Any = [
msg_content for msg in new_messages for msg_content in msg.content
]
if all(
isinstance(content, openai.types.beta.threads.MessageContentText)
for content in answer
):
answer = "\n".join(content.text.value for content in answer)
return OpenAIAssistantFinish(
return_values={"output": answer},
log="",
run_id=run_id,
thread_id=thread_id,
)
elif run.status == "requires_action":
if not self.as_agent:
return run.required_action.submit_tool_outputs.tool_calls
actions = []
for tool_call in run.required_action.submit_tool_outputs.tool_calls:
function = tool_call.function
args = json.loads(function.arguments)
if len(args) == 1 and "__arg1" in args:
args = args["__arg1"]
actions.append(
OpenAIAssistantAction(
tool=function.name,
tool_input=args,
tool_call_id=tool_call.id,
log="",
run_id=run_id,
thread_id=thread_id,
)
)
return actions
else:
run_info = json.dumps(run.dict(), indent=2)
raise ValueError(
f"Unexpected run status: {run.status}. Full run info:\n\n{run_info})"
)
def _wait_for_run(self, run_id: str, thread_id: str) -> Any:
in_progress = True
while in_progress:
run = self.client.beta.threads.runs.retrieve(run_id, thread_id=thread_id)
in_progress = run.status in ("in_progress", "queued")
if in_progress:
sleep(self.check_every_ms / 1000)
return run
OpenAIAssistantRunnable,
)

3
libs/langchain/langchain/agents/openai_assistant/__init__.py Normal file
View File

@ -0,0 +1,3 @@
from langchain_experimental.openai_assistant.base import OpenAIAssistantRunnable
__all__ = ["OpenAIAssistantRunnable"]

376
libs/langchain/langchain/agents/openai_assistant/base.py Normal file
View File

@ -0,0 +1,376 @@
from __future__ import annotations
import json
from time import sleep
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Sequence, Tuple, Union
from langchain.callbacks.manager import CallbackManager
from langchain.load import dumpd
from langchain.pydantic_v1 import Field
from langchain.schema.agent import AgentAction, AgentFinish
from langchain.schema.runnable import RunnableConfig, RunnableSerializable
from langchain.tools.base import BaseTool
from langchain.tools.render import format_tool_to_openai_tool
if TYPE_CHECKING:
import openai
from openai.types.beta.threads import ThreadMessage
from openai.types.beta.threads.required_action_function_tool_call import (
RequiredActionFunctionToolCall,
)
class OpenAIAssistantFinish(AgentFinish):
"""AgentFinish with run and thread metadata."""
run_id: str
thread_id: str
class OpenAIAssistantAction(AgentAction):
"""AgentAction with info needed to submit custom tool output to existing run."""
tool_call_id: str
run_id: str
thread_id: str
def _get_openai_client() -> openai.OpenAI:
try:
import openai
return openai.OpenAI()
except ImportError as e:
raise ImportError(
"Unable to import openai, please install with `pip install openai`."
) from e
except AttributeError as e:
raise AttributeError(
"Please make sure you are using a v1.1-compatible version of openai. You "
'can install with `pip install "openai>=1.1"`.'
) from e
OutputType = Union[
List[OpenAIAssistantAction],
OpenAIAssistantFinish,
List["ThreadMessage"],
List["RequiredActionFunctionToolCall"],
]
class OpenAIAssistantRunnable(RunnableSerializable[Dict, OutputType]):
"""Run an OpenAI Assistant.
Example using OpenAI tools:
.. code-block:: python
from langchain_experimental.openai_assistant import OpenAIAssistantRunnable
interpreter_assistant = OpenAIAssistantRunnable.create_assistant(
name="langchain assistant",
instructions="You are a personal math tutor. Write and run code to answer math questions.",
tools=[{"type": "code_interpreter"}],
model="gpt-4-1106-preview"
)
output = interpreter_assistant.invoke({"content": "What's 10 - 4 raised to the 2.7"})
Example using custom tools and AgentExecutor:
.. code-block:: python
from langchain_experimental.openai_assistant import OpenAIAssistantRunnable
from langchain.agents import AgentExecutor
from langchain.tools import E2BDataAnalysisTool
tools = [E2BDataAnalysisTool(api_key="...")]
agent = OpenAIAssistantRunnable.create_assistant(
name="langchain assistant e2b tool",
instructions="You are a personal math tutor. Write and run code to answer math questions.",
tools=tools,
model="gpt-4-1106-preview",
as_agent=True
)
agent_executor = AgentExecutor(agent=agent, tools=tools)
agent_executor.invoke({"content": "What's 10 - 4 raised to the 2.7"})
Example using custom tools and custom execution:
.. code-block:: python
from langchain_experimental.openai_assistant import OpenAIAssistantRunnable
from langchain.agents import AgentExecutor
from langchain.schema.agent import AgentFinish
from langchain.tools import E2BDataAnalysisTool
tools = [E2BDataAnalysisTool(api_key="...")]
agent = OpenAIAssistantRunnable.create_assistant(
name="langchain assistant e2b tool",
instructions="You are a personal math tutor. Write and run code to answer math questions.",
tools=tools,
model="gpt-4-1106-preview",
as_agent=True
)
def execute_agent(agent, tools, input):
tool_map = {tool.name: tool for tool in tools}
response = agent.invoke(input)
while not isinstance(response, AgentFinish):
tool_outputs = []
for action in response:
tool_output = tool_map[action.tool].invoke(action.tool_input)
tool_outputs.append({"output": tool_output, "tool_call_id": action.tool_call_id})
response = agent.invoke(
{
"tool_outputs": tool_outputs,
"run_id": action.run_id,
"thread_id": action.thread_id
}
)
return response
response = execute_agent(agent, tools, {"content": "What's 10 - 4 raised to the 2.7"})
next_response = execute_agent(agent, tools, {"content": "now add 17.241", "thread_id": response.thread_id})
""" # noqa: E501
client: openai.OpenAI = Field(default_factory=_get_openai_client)
"""OpenAI client."""
assistant_id: str
"""OpenAI assistant id."""
check_every_ms: float = 1_000.0
"""Frequency with which to check run progress in ms."""
as_agent: bool = False
"""Use as a LangChain agent, compatible with the AgentExecutor."""
@classmethod
def create_assistant(
cls,
name: str,
instructions: str,
tools: Sequence[Union[BaseTool, dict]],
model: str,
*,
client: Optional[openai.OpenAI] = None,
**kwargs: Any,
) -> OpenAIAssistantRunnable:
"""Create an OpenAI Assistant and instantiate the Runnable.
Args:
name: Assistant name.
instructions: Assistant instructions.
tools: Assistant tools. Can be passed in in OpenAI format or as BaseTools.
model: Assistant model to use.
client: OpenAI client. Will create default client if not specified.
Returns:
OpenAIAssistantRunnable configured to run using the created assistant.
"""
client = client or _get_openai_client()
openai_tools: List = []
for tool in tools:
oai_tool = (
tool if isinstance(tool, dict) else format_tool_to_openai_tool(tool)
)
openai_tools.append(oai_tool)
assistant = client.beta.assistants.create(
name=name,
instructions=instructions,
tools=openai_tools,
model=model,
)
return cls(assistant_id=assistant.id, **kwargs)
def invoke(
self, input: dict, config: Optional[RunnableConfig] = None
) -> OutputType:
"""Invoke assistant.
Args:
input: Runnable input dict that can have:
content: User message when starting a new run.
thread_id: Existing thread to use.
run_id: Existing run to use. Should only be supplied when providing
the tool output for a required action after an initial invocation.
file_ids: File ids to include in new run. Used for retrieval.
message_metadata: Metadata to associate with new message.
thread_metadata: Metadata to associate with new thread. Only relevant
when new thread being created.
instructions: Additional run instructions.
model: Override Assistant model for this run.
tools: Override Assistant tools for this run.
run_metadata: Metadata to associate with new run.
config: Runnable config:
Return:
If self.as_agent, will return
Union[List[OpenAIAssistantAction], OpenAIAssistantFinish]. Otherwise
will return OpenAI types
Union[List[ThreadMessage], List[RequiredActionFunctionToolCall]].
"""
config = config or {}
callback_manager = CallbackManager.configure(
inheritable_callbacks=config.get("callbacks"),
inheritable_tags=config.get("tags"),
inheritable_metadata=config.get("metadata"),
)
run_manager = callback_manager.on_chain_start(
dumpd(self), input, name=config.get("run_name")
)
try:
# Being run within AgentExecutor and there are tool outputs to submit.
if self.as_agent and input.get("intermediate_steps"):
tool_outputs = self._parse_intermediate_steps(
input["intermediate_steps"]
)
run = self.client.beta.threads.runs.submit_tool_outputs(**tool_outputs)
# Starting a new thread and a new run.
elif "thread_id" not in input:
thread = {
"messages": [
{
"role": "user",
"content": input["content"],
"file_ids": input.get("file_ids", []),
"metadata": input.get("message_metadata"),
}
],
"metadata": input.get("thread_metadata"),
}
run = self._create_thread_and_run(input, thread)
# Starting a new run in an existing thread.
elif "run_id" not in input:
_ = self.client.beta.threads.messages.create(
input["thread_id"],
content=input["content"],
role="user",
file_ids=input.get("file_ids", []),
metadata=input.get("message_metadata"),
)
run = self._create_run(input)
# Submitting tool outputs to an existing run, outside the AgentExecutor
# framework.
else:
run = self.client.beta.threads.runs.submit_tool_outputs(**input)
run = self._wait_for_run(run.id, run.thread_id)
except BaseException as e:
run_manager.on_chain_error(e)
raise e
try:
response = self._get_response(run)
except BaseException as e:
run_manager.on_chain_error(e, metadata=run.dict())
raise e
else:
run_manager.on_chain_end(response)
return response
def _parse_intermediate_steps(
self, intermediate_steps: List[Tuple[OpenAIAssistantAction, str]]
) -> dict:
last_action, last_output = intermediate_steps[-1]
run = self._wait_for_run(last_action.run_id, last_action.thread_id)
required_tool_call_ids = {
tc.id for tc in run.required_action.submit_tool_outputs.tool_calls
}
tool_outputs = [
{"output": output, "tool_call_id": action.tool_call_id}
for action, output in intermediate_steps
if action.tool_call_id in required_tool_call_ids
]
submit_tool_outputs = {
"tool_outputs": tool_outputs,
"run_id": last_action.run_id,
"thread_id": last_action.thread_id,
}
return submit_tool_outputs
def _create_run(self, input: dict) -> Any:
params = {
k: v
for k, v in input.items()
if k in ("instructions", "model", "tools", "run_metadata")
}
return self.client.beta.threads.runs.create(
input["thread_id"],
assistant_id=self.assistant_id,
**params,
)
def _create_thread_and_run(self, input: dict, thread: dict) -> Any:
params = {
k: v
for k, v in input.items()
if k in ("instructions", "model", "tools", "run_metadata")
}
run = self.client.beta.threads.create_and_run(
assistant_id=self.assistant_id,
thread=thread,
**params,
)
return run
def _get_response(self, run: Any) -> Any:
# TODO: Pagination
if run.status == "completed":
import openai
messages = self.client.beta.threads.messages.list(
run.thread_id, order="asc"
)
new_messages = [msg for msg in messages if msg.run_id == run.id]
if not self.as_agent:
return new_messages
answer: Any = [
msg_content for msg in new_messages for msg_content in msg.content
]
if all(
isinstance(content, openai.types.beta.threads.MessageContentText)
for content in answer
):
answer = "\n".join(content.text.value for content in answer)
return OpenAIAssistantFinish(
return_values={"output": answer},
log="",
run_id=run.id,
thread_id=run.thread_id,
)
elif run.status == "requires_action":
if not self.as_agent:
return run.required_action.submit_tool_outputs.tool_calls
actions = []
for tool_call in run.required_action.submit_tool_outputs.tool_calls:
function = tool_call.function
args = json.loads(function.arguments)
if len(args) == 1 and "__arg1" in args:
args = args["__arg1"]
actions.append(
OpenAIAssistantAction(
tool=function.name,
tool_input=args,
tool_call_id=tool_call.id,
log="",
run_id=run.id,
thread_id=run.thread_id,
)
)
return actions
else:
run_info = json.dumps(run.dict(), indent=2)
raise ValueError(
f"Unexpected run status: {run.status}. Full run info:\n\n{run_info})"
)
def _wait_for_run(self, run_id: str, thread_id: str) -> Any:
in_progress = True
while in_progress:
run = self.client.beta.threads.runs.retrieve(run_id, thread_id=thread_id)
in_progress = run.status in ("in_progress", "queued")
if in_progress:
sleep(self.check_every_ms / 1000)
return run