ai-forever/gigachain

Ошибка при запуске агента с функциями на базе GigaChat

gritaro opened this issue · 2 comments

gritaro commented

Description: Ubuntu 22.04.3 LTS
Python 3.10.12
gigachain 0.1.4
gigachain-community 0.0.16
gigachain-core 0.1.17
gigachat 0.1.16

@Rai220 здравствуйте
Пытаюсь запускать код из примера https://github.com/ai-forever/gigachain/blob/master/docs/docs/modules/agents/how_to/gigachat_agent.ipynb

from langchain_community.chat_models import GigaChat
from langchain.tools import tool
import pyfiglet
from langchain.agents import AgentExecutor, create_gigachat_functions_agent
from langchain.schema import HumanMessage, SystemMessage, AIMessage

@tool
def draw_bunner(number: str) -> None:
    """Рисует баннер с текстом

    Args:
        number (str): Число, который нужно нарисовать на баннере
    """
    pyfiglet.print_figlet(number, font="epic")

giga = GigaChat(credentials=***", verify_ssl_certs=False, model="GigaChat-Plus")
new_tools = [draw_bunner]
agent = create_gigachat_functions_agent(giga, new_tools)
agent_executor = AgentExecutor(
    agent=agent,
    tools=new_tools,
    verbose=True,
)

agent_executor.invoke(
    {"input": "Найди курс биткоина в долларах и нарисуй это число на банере."}
)["output"]

В ответ получаю HTTP 422
Пробовал указывать разные модели, результат один
Стектрейс:

> Entering new AgentExecutor chain...
Traceback (most recent call last):
  File "/home/roman/GIT/hass/gigachain/1test-model.py", line 25, in <module>
    agent_executor.invoke(
  File "/usr/local/lib/python3.10/dist-packages/langchain/chains/base.py", line 162, in invoke
    raise e
  File "/usr/local/lib/python3.10/dist-packages/langchain/chains/base.py", line 156, in invoke
    self._call(inputs, run_manager=run_manager)
  File "/usr/local/lib/python3.10/dist-packages/langchain/agents/agent.py", line 1391, in _call
    next_step_output = self._take_next_step(
  File "/usr/local/lib/python3.10/dist-packages/langchain/agents/agent.py", line 1097, in _take_next_step
    [
  File "/usr/local/lib/python3.10/dist-packages/langchain/agents/agent.py", line 1097, in <listcomp>
    [
  File "/usr/local/lib/python3.10/dist-packages/langchain/agents/agent.py", line 1125, in _iter_next_step
    output = self.agent.plan(
  File "/usr/local/lib/python3.10/dist-packages/langchain/agents/agent.py", line 387, in plan
    for chunk in self.runnable.stream(inputs, config={"callbacks": callbacks}):
  File "/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/base.py", line 2424, in stream
    yield from self.transform(iter([input]), config, **kwargs)
  File "/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/base.py", line 2411, in transform
    yield from self._transform_stream_with_config(
  File "/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/base.py", line 1497, in _transform_stream_with_config
    chunk: Output = context.run(next, iterator)  # type: ignore
  File "/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/base.py", line 2375, in _transform
    for output in final_pipeline:
  File "/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/base.py", line 1035, in transform
    for chunk in input:
  File "/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/base.py", line 4145, in transform
    yield from self.bound.transform(
  File "/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/base.py", line 1045, in transform
    yield from self.stream(final, config, **kwargs)
  File "/usr/local/lib/python3.10/dist-packages/langchain_core/language_models/chat_models.py", line 250, in stream
    raise e
  File "/usr/local/lib/python3.10/dist-packages/langchain_core/language_models/chat_models.py", line 234, in stream
    for chunk in self._stream(
  File "/usr/local/lib/python3.10/dist-packages/langchain_community/chat_models/gigachat.py", line 235, in _stream
    for chunk in self._client.stream(payload):
  File "/usr/local/lib/python3.10/dist-packages/gigachat/client.py", line 246, in stream
    for chunk in stream_chat.sync(self._client, chat=chat, access_token=self.token):
  File "/usr/local/lib/python3.10/dist-packages/gigachat/api/stream_chat.py", line 90, in sync
    _check_response(response)
  File "/usr/local/lib/python3.10/dist-packages/gigachat/api/stream_chat.py", line 79, in _check_response
    raise ResponseError(response.url, response.status_code, b"", response.headers)
gigachat.exceptions.ResponseError: (URL('https://gigachat.devices.sberbank.ru/api/v1/chat/completions'), 422, b'', Headers([('server', 'nginx'), ('date', 'Fri, 16 Feb 2024 17:33:26 GMT'), ('content-type', 'application/json; charset=utf-8'), ('content-length', '70'), ('connection', 'keep-alive'), ('access-control-allow-credentials', 'true'), ('access-control-allow-headers', 'Origin, X-Requested-With, Content-Type, Accept, Authorization'), ('access-control-allow-methods', 'GET, POST, DELETE, OPTIONS'), ('access-control-allow-origin', 'https://beta.saluteai.sberdevices.ru'), ('x-request-id', 'b05d648f-ac12-4e13-86b3-dfcc782ed7bd'), ('x-session-id', '625022d4-55a7-417e-86e8-47432f94efd6'), ('allow', 'GET, POST'), ('strict-transport-security', 'max-age=31536000; includeSubDomains'), ('allow', 'GET, POST'), ('strict-transport-security', 'max-age=31536000; includeSubDomains')]))
Rai220 commented

Здравствуйте! Нужен доступ именно к "модели с поддержкой функций", сейчас она есть только у ограниченного круга лиц в рамках тестирования. В ближайшее время будет доступна для всех. Если очень нужна до запуска, можно обратиться gigachat@sberbank.ru, вероятно помогут получить к ней доступ (но тут я не могу гарантировать).

  • 👍1