INVALID_TOOL_RESULTS
You are passing too many, too few, or mismatched ToolMessages to a model.
When using a model to call tools, the AIMessage
the model responds with will contain a tool_calls array. To continue the flow, the next messages you pass back to the model must
be exactly one ToolMessage for each item in that array containing the result of that tool call. Each ToolMessage must have a tool_call_id field
that matches one of the tool_calls on the AIMessage.
For example, given the following response from a model:
from typing import List
from langchain_core.messages import BaseMessage, HumanMessage
from langchain_core.tools import tool
from langchain_openai import ChatOpenAI
model = ChatOpenAI(model="gpt-4o-mini")
@tool
def foo_tool() -> str:
"""
A dummy tool that returns 'action complete!'
"""
return "action complete!"
model_with_tools = model.bind_tools([foo_tool])
chat_history: List[BaseMessage] = [
HumanMessage(content='Call tool "foo" twice with no arguments')
]
response_message = model_with_tools.invoke(chat_history)
print(response_message.tool_calls)
[{'name': 'foo_tool', 'args': {}, 'id': 'call_dq9O0eGHrryBwDRCnk0deHK4', 'type': 'tool_call'}, {'name': 'foo_tool', 'args': {}, 'id': 'call_mjLuNyXNHoUIXHiBtXhaWdxN', 'type': 'tool_call'}]
Calling the model with only one tool response would result in an error:
from langchain_core.messages import AIMessage, ToolMessage
tool_call = response_message.tool_calls[0]
tool_response = foo_tool.invoke(tool_call)
chat_history.append(
AIMessage(
content=response_message.content,
additional_kwargs=response_message.additional_kwargs,
)
)
chat_history.append(
ToolMessage(content=str(tool_response), tool_call_id=tool_call.get("id"))
)
final_response = model_with_tools.invoke(chat_history)
print(final_response)
API Reference:AIMessage | ToolMessage
---------------------------------------------------------------------------
``````output
BadRequestError Traceback (most recent call last)
``````output
Cell In[3], line 9
6 chat_history.append(AIMessage(content=response_message.content, additional_kwargs=response_message.additional_kwargs))
7 chat_history.append(ToolMessage(content=str(tool_response), tool_call_id=tool_call.get('id')))
----> 9 final_response = model_with_tools.invoke(chat_history)
10 print(final_response)
``````output
File ~/langchain/oss-py/libs/core/langchain_core/runnables/base.py:5354, in RunnableBindingBase.invoke(self, input, config, **kwargs)
5348 def invoke(
5349 self,
5350 input: Input,
5351 config: Optional[RunnableConfig] = None,
5352 **kwargs: Optional[Any],
5353 ) -> Output:
-> 5354 return self.bound.invoke(
5355 input,
5356 self._merge_configs(config),
5357 **{**self.kwargs, **kwargs},
5358 )
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:286, in BaseChatModel.invoke(self, input, config, stop, **kwargs)
275 def invoke(
276 self,
277 input: LanguageModelInput,
(...)
281 **kwargs: Any,
282 ) -> BaseMessage:
283 config = ensure_config(config)
284 return cast(
285 ChatGeneration,
--> 286 self.generate_prompt(
287 [self._convert_input(input)],
288 stop=stop,
289 callbacks=config.get("callbacks"),
290 tags=config.get("tags"),
291 metadata=config.get("metadata"),
292 run_name=config.get("run_name"),
293 run_id=config.pop("run_id", None),
294 **kwargs,
295 ).generations[0][0],
296 ).message
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:786, in BaseChatModel.generate_prompt(self, prompts, stop, callbacks, **kwargs)
778 def generate_prompt(
779 self,
780 prompts: list[PromptValue],
(...)
783 **kwargs: Any,
784 ) -> LLMResult:
785 prompt_messages = [p.to_messages() for p in prompts]
--> 786 return self.generate(prompt_messages, stop=stop, callbacks=callbacks, **kwargs)
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:643, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, run_id, **kwargs)
641 if run_managers:
642 run_managers[i].on_llm_error(e, response=LLMResult(generations=[]))
--> 643 raise e
644 flattened_outputs = [
645 LLMResult(generations=[res.generations], llm_output=res.llm_output) # type: ignore[list-item]
646 for res in results
647 ]
648 llm_output = self._combine_llm_outputs([res.llm_output for res in results])
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:633, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, run_id, **kwargs)
630 for i, m in enumerate(messages):
631 try:
632 results.append(
--> 633 self._generate_with_cache(
634 m,
635 stop=stop,
636 run_manager=run_managers[i] if run_managers else None,
637 **kwargs,
638 )
639 )
640 except BaseException as e:
641 if run_managers:
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:851, in BaseChatModel._generate_with_cache(self, messages, stop, run_manager, **kwargs)
849 else:
850 if inspect.signature(self._generate).parameters.get("run_manager"):
--> 851 result = self._generate(
852 messages, stop=stop, run_manager=run_manager, **kwargs
853 )
854 else:
855 result = self._generate(messages, stop=stop, **kwargs)
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/langchain_openai/chat_models/base.py:686, in BaseChatOpenAI._generate(self, messages, stop, run_manager, **kwargs)
684 generation_info = {"headers": dict(raw_response.headers)}
685 else:
--> 686 response = self.client.create(**payload)
687 return self._create_chat_result(response, generation_info)
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_utils/_utils.py:274, in required_args.<locals>.inner.<locals>.wrapper(*args, **kwargs)
272 msg = f"Missing required argument: {quote(missing[0])}"
273 raise TypeError(msg)
--> 274 return func(*args, **kwargs)
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/resources/chat/completions.py:742, in Completions.create(self, messages, model, frequency_penalty, function_call, functions, logit_bias, logprobs, max_completion_tokens, max_tokens, metadata, n, parallel_tool_calls, presence_penalty, response_format, seed, service_tier, stop, store, stream, stream_options, temperature, tool_choice, tools, top_logprobs, top_p, user, extra_headers, extra_query, extra_body, timeout)
704 @required_args(["messages", "model"], ["messages", "model", "stream"])
705 def create(
706 self,
(...)
739 timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
740 ) -> ChatCompletion | Stream[ChatCompletionChunk]:
741 validate_response_format(response_format)
--> 742 return self._post(
743 "/chat/completions",
744 body=maybe_transform(
745 {
746 "messages": messages,
747 "model": model,
748 "frequency_penalty": frequency_penalty,
749 "function_call": function_call,
750 "functions": functions,
751 "logit_bias": logit_bias,
752 "logprobs": logprobs,
753 "max_completion_tokens": max_completion_tokens,
754 "max_tokens": max_tokens,
755 "metadata": metadata,
756 "n": n,
757 "parallel_tool_calls": parallel_tool_calls,
758 "presence_penalty": presence_penalty,
759 "response_format": response_format,
760 "seed": seed,
761 "service_tier": service_tier,
762 "stop": stop,
763 "store": store,
764 "stream": stream,
765 "stream_options": stream_options,
766 "temperature": temperature,
767 "tool_choice": tool_choice,
768 "tools": tools,
769 "top_logprobs": top_logprobs,
770 "top_p": top_p,
771 "user": user,
772 },
773 completion_create_params.CompletionCreateParams,
774 ),
775 options=make_request_options(
776 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
777 ),
778 cast_to=ChatCompletion,
779 stream=stream or False,
780 stream_cls=Stream[ChatCompletionChunk],
781 )
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_base_client.py:1270, in SyncAPIClient.post(self, path, cast_to, body, options, files, stream, stream_cls)
1256 def post(
1257 self,
1258 path: str,
(...)
1265 stream_cls: type[_StreamT] | None = None,
1266 ) -> ResponseT | _StreamT:
1267 opts = FinalRequestOptions.construct(
1268 method="post", url=path, json_data=body, files=to_httpx_files(files), **options
1269 )
-> 1270 return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_base_client.py:947, in SyncAPIClient.request(self, cast_to, options, remaining_retries, stream, stream_cls)
944 else:
945 retries_taken = 0
--> 947 return self._request(
948 cast_to=cast_to,
949 options=options,
950 stream=stream,
951 stream_cls=stream_cls,
952 retries_taken=retries_taken,
953 )
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_base_client.py:1051, in SyncAPIClient._request(self, cast_to, options, retries_taken, stream, stream_cls)
1048 err.response.read()
1050 log.debug("Re-raising status error")
-> 1051 raise self._make_status_error_from_response(err.response) from None
1053 return self._process_response(
1054 cast_to=cast_to,
1055 options=options,
(...)
1059 retries_taken=retries_taken,
1060 )
``````output
BadRequestError: Error code: 400 - {'error': {'message': "An assistant message with 'tool_calls' must be followed by tool messages responding to each 'tool_call_id'. The following tool_call_ids did not have response messages: call_mjLuNyXNHoUIXHiBtXhaWdxN", 'type': 'invalid_request_error', 'param': 'messages', 'code': None}}
If we add a second response, the call will succeed as expected because we now have one tool response per tool call:
tool_response_2 = foo_tool.invoke(response_message.tool_calls[1])
chat_history.append(tool_response_2)
model_with_tools.invoke(chat_history)
AIMessage(content='Both calls to the tool "foo" have been completed successfully. The output for each call is "action complete!".', additional_kwargs={'refusal': None}, response_metadata={'token_usage': {'completion_tokens': 24, 'prompt_tokens': 137, 'total_tokens': 161, 'completion_tokens_details': {'audio_tokens': None, 'reasoning_tokens': 0}, 'prompt_tokens_details': {'audio_tokens': None, 'cached_tokens': 0}}, 'model_name': 'gpt-4o-mini-2024-07-18', 'system_fingerprint': 'fp_e2bde53e6e', 'finish_reason': 'stop', 'logprobs': None}, id='run-b5ac3c54-4e26-4da4-853a-d0ab1cba90e0-0', usage_metadata={'input_tokens': 137, 'output_tokens': 24, 'total_tokens': 161, 'input_token_details': {'cache_read': 0}, 'output_token_details': {'reasoning': 0}})
But if we add a duplicate, extra tool response, the call will fail again:
duplicate_tool_response_2 = foo_tool.invoke(response_message.tool_calls[1])
chat_history.append(duplicate_tool_response_2)
await model_with_tools.invoke(chat_history)
---------------------------------------------------------------------------
``````output
BadRequestError Traceback (most recent call last)
``````output
Cell In[7], line 5
1 duplicate_tool_response_2 = foo_tool.invoke(response_message.tool_calls[1])
3 chat_history.append(duplicate_tool_response_2)
----> 5 await model_with_tools.invoke(chat_history)
``````output
File ~/langchain/oss-py/libs/core/langchain_core/runnables/base.py:5354, in RunnableBindingBase.invoke(self, input, config, **kwargs)
5348 def invoke(
5349 self,
5350 input: Input,
5351 config: Optional[RunnableConfig] = None,
5352 **kwargs: Optional[Any],
5353 ) -> Output:
-> 5354 return self.bound.invoke(
5355 input,
5356 self._merge_configs(config),
5357 **{**self.kwargs, **kwargs},
5358 )
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:286, in BaseChatModel.invoke(self, input, config, stop, **kwargs)
275 def invoke(
276 self,
277 input: LanguageModelInput,
(...)
281 **kwargs: Any,
282 ) -> BaseMessage:
283 config = ensure_config(config)
284 return cast(
285 ChatGeneration,
--> 286 self.generate_prompt(
287 [self._convert_input(input)],
288 stop=stop,
289 callbacks=config.get("callbacks"),
290 tags=config.get("tags"),
291 metadata=config.get("metadata"),
292 run_name=config.get("run_name"),
293 run_id=config.pop("run_id", None),
294 **kwargs,
295 ).generations[0][0],
296 ).message
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:786, in BaseChatModel.generate_prompt(self, prompts, stop, callbacks, **kwargs)
778 def generate_prompt(
779 self,
780 prompts: list[PromptValue],
(...)
783 **kwargs: Any,
784 ) -> LLMResult:
785 prompt_messages = [p.to_messages() for p in prompts]
--> 786 return self.generate(prompt_messages, stop=stop, callbacks=callbacks, **kwargs)
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:643, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, run_id, **kwargs)
641 if run_managers:
642 run_managers[i].on_llm_error(e, response=LLMResult(generations=[]))
--> 643 raise e
644 flattened_outputs = [
645 LLMResult(generations=[res.generations], llm_output=res.llm_output) # type: ignore[list-item]
646 for res in results
647 ]
648 llm_output = self._combine_llm_outputs([res.llm_output for res in results])
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:633, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, run_id, **kwargs)
630 for i, m in enumerate(messages):
631 try:
632 results.append(
--> 633 self._generate_with_cache(
634 m,
635 stop=stop,
636 run_manager=run_managers[i] if run_managers else None,
637 **kwargs,
638 )
639 )
640 except BaseException as e:
641 if run_managers:
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:851, in BaseChatModel._generate_with_cache(self, messages, stop, run_manager, **kwargs)
849 else:
850 if inspect.signature(self._generate).parameters.get("run_manager"):
--> 851 result = self._generate(
852 messages, stop=stop, run_manager=run_manager, **kwargs
853 )
854 else:
855 result = self._generate(messages, stop=stop, **kwargs)
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/langchain_openai/chat_models/base.py:686, in BaseChatOpenAI._generate(self, messages, stop, run_manager, **kwargs)
684 generation_info = {"headers": dict(raw_response.headers)}
685 else:
--> 686 response = self.client.create(**payload)
687 return self._create_chat_result(response, generation_info)
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_utils/_utils.py:274, in required_args.<locals>.inner.<locals>.wrapper(*args, **kwargs)
272 msg = f"Missing required argument: {quote(missing[0])}"
273 raise TypeError(msg)
--> 274 return func(*args, **kwargs)
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/resources/chat/completions.py:742, in Completions.create(self, messages, model, frequency_penalty, function_call, functions, logit_bias, logprobs, max_completion_tokens, max_tokens, metadata, n, parallel_tool_calls, presence_penalty, response_format, seed, service_tier, stop, store, stream, stream_options, temperature, tool_choice, tools, top_logprobs, top_p, user, extra_headers, extra_query, extra_body, timeout)
704 @required_args(["messages", "model"], ["messages", "model", "stream"])
705 def create(
706 self,
(...)
739 timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
740 ) -> ChatCompletion | Stream[ChatCompletionChunk]:
741 validate_response_format(response_format)
--> 742 return self._post(
743 "/chat/completions",
744 body=maybe_transform(
745 {
746 "messages": messages,
747 "model": model,
748 "frequency_penalty": frequency_penalty,
749 "function_call": function_call,
750 "functions": functions,
751 "logit_bias": logit_bias,
752 "logprobs": logprobs,
753 "max_completion_tokens": max_completion_tokens,
754 "max_tokens": max_tokens,
755 "metadata": metadata,
756 "n": n,
757 "parallel_tool_calls": parallel_tool_calls,
758 "presence_penalty": presence_penalty,
759 "response_format": response_format,
760 "seed": seed,
761 "service_tier": service_tier,
762 "stop": stop,
763 "store": store,
764 "stream": stream,
765 "stream_options": stream_options,
766 "temperature": temperature,
767 "tool_choice": tool_choice,
768 "tools": tools,
769 "top_logprobs": top_logprobs,
770 "top_p": top_p,
771 "user": user,
772 },
773 completion_create_params.CompletionCreateParams,
774 ),
775 options=make_request_options(
776 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
777 ),
778 cast_to=ChatCompletion,
779 stream=stream or False,
780 stream_cls=Stream[ChatCompletionChunk],
781 )
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_base_client.py:1270, in SyncAPIClient.post(self, path, cast_to, body, options, files, stream, stream_cls)
1256 def post(
1257 self,
1258 path: str,
(...)
1265 stream_cls: type[_StreamT] | None = None,
1266 ) -> ResponseT | _StreamT:
1267 opts = FinalRequestOptions.construct(
1268 method="post", url=path, json_data=body, files=to_httpx_files(files), **options
1269 )
-> 1270 return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_base_client.py:947, in SyncAPIClient.request(self, cast_to, options, remaining_retries, stream, stream_cls)
944 else:
945 retries_taken = 0
--> 947 return self._request(
948 cast_to=cast_to,
949 options=options,
950 stream=stream,
951 stream_cls=stream_cls,
952 retries_taken=retries_taken,
953 )
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_base_client.py:1051, in SyncAPIClient._request(self, cast_to, options, retries_taken, stream, stream_cls)
1048 err.response.read()
1050 log.debug("Re-raising status error")
-> 1051 raise self._make_status_error_from_response(err.response) from None
1053 return self._process_response(
1054 cast_to=cast_to,
1055 options=options,
(...)
1059 retries_taken=retries_taken,
1060 )
``````output
BadRequestError: Error code: 400 - {'error': {'message': "Invalid parameter: messages with role 'tool' must be a response to a preceeding message with 'tool_calls'.", 'type': 'invalid_request_error', 'param': 'messages.[4].role', 'code': None}}
You should additionally not pass ToolMessages back to to a model if they are not preceded by an AIMessage with tool calls. For example, this will fail:
model_with_tools.invoke(
[ToolMessage(content="action completed!", tool_call_id="dummy")]
)
---------------------------------------------------------------------------
``````output
BadRequestError Traceback (most recent call last)
``````output
Cell In[8], line 1
----> 1 model_with_tools.invoke([ToolMessage(content="action completed!", tool_call_id="dummy")])
``````output
File ~/langchain/oss-py/libs/core/langchain_core/runnables/base.py:5354, in RunnableBindingBase.invoke(self, input, config, **kwargs)
5348 def invoke(
5349 self,
5350 input: Input,
5351 config: Optional[RunnableConfig] = None,
5352 **kwargs: Optional[Any],
5353 ) -> Output:
-> 5354 return self.bound.invoke(
5355 input,
5356 self._merge_configs(config),
5357 **{**self.kwargs, **kwargs},
5358 )
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:286, in BaseChatModel.invoke(self, input, config, stop, **kwargs)
275 def invoke(
276 self,
277 input: LanguageModelInput,
(...)
281 **kwargs: Any,
282 ) -> BaseMessage:
283 config = ensure_config(config)
284 return cast(
285 ChatGeneration,
--> 286 self.generate_prompt(
287 [self._convert_input(input)],
288 stop=stop,
289 callbacks=config.get("callbacks"),
290 tags=config.get("tags"),
291 metadata=config.get("metadata"),
292 run_name=config.get("run_name"),
293 run_id=config.pop("run_id", None),
294 **kwargs,
295 ).generations[0][0],
296 ).message
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:786, in BaseChatModel.generate_prompt(self, prompts, stop, callbacks, **kwargs)
778 def generate_prompt(
779 self,
780 prompts: list[PromptValue],
(...)
783 **kwargs: Any,
784 ) -> LLMResult:
785 prompt_messages = [p.to_messages() for p in prompts]
--> 786 return self.generate(prompt_messages, stop=stop, callbacks=callbacks, **kwargs)
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:643, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, run_id, **kwargs)
641 if run_managers:
642 run_managers[i].on_llm_error(e, response=LLMResult(generations=[]))
--> 643 raise e
644 flattened_outputs = [
645 LLMResult(generations=[res.generations], llm_output=res.llm_output) # type: ignore[list-item]
646 for res in results
647 ]
648 llm_output = self._combine_llm_outputs([res.llm_output for res in results])
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:633, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, run_id, **kwargs)
630 for i, m in enumerate(messages):
631 try:
632 results.append(
--> 633 self._generate_with_cache(
634 m,
635 stop=stop,
636 run_manager=run_managers[i] if run_managers else None,
637 **kwargs,
638 )
639 )
640 except BaseException as e:
641 if run_managers:
``````output
File ~/langchain/oss-py/libs/core/langchain_core/language_models/chat_models.py:851, in BaseChatModel._generate_with_cache(self, messages, stop, run_manager, **kwargs)
849 else:
850 if inspect.signature(self._generate).parameters.get("run_manager"):
--> 851 result = self._generate(
852 messages, stop=stop, run_manager=run_manager, **kwargs
853 )
854 else:
855 result = self._generate(messages, stop=stop, **kwargs)
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/langchain_openai/chat_models/base.py:686, in BaseChatOpenAI._generate(self, messages, stop, run_manager, **kwargs)
684 generation_info = {"headers": dict(raw_response.headers)}
685 else:
--> 686 response = self.client.create(**payload)
687 return self._create_chat_result(response, generation_info)
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_utils/_utils.py:274, in required_args.<locals>.inner.<locals>.wrapper(*args, **kwargs)
272 msg = f"Missing required argument: {quote(missing[0])}"
273 raise TypeError(msg)
--> 274 return func(*args, **kwargs)
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/resources/chat/completions.py:742, in Completions.create(self, messages, model, frequency_penalty, function_call, functions, logit_bias, logprobs, max_completion_tokens, max_tokens, metadata, n, parallel_tool_calls, presence_penalty, response_format, seed, service_tier, stop, store, stream, stream_options, temperature, tool_choice, tools, top_logprobs, top_p, user, extra_headers, extra_query, extra_body, timeout)
704 @required_args(["messages", "model"], ["messages", "model", "stream"])
705 def create(
706 self,
(...)
739 timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
740 ) -> ChatCompletion | Stream[ChatCompletionChunk]:
741 validate_response_format(response_format)
--> 742 return self._post(
743 "/chat/completions",
744 body=maybe_transform(
745 {
746 "messages": messages,
747 "model": model,
748 "frequency_penalty": frequency_penalty,
749 "function_call": function_call,
750 "functions": functions,
751 "logit_bias": logit_bias,
752 "logprobs": logprobs,
753 "max_completion_tokens": max_completion_tokens,
754 "max_tokens": max_tokens,
755 "metadata": metadata,
756 "n": n,
757 "parallel_tool_calls": parallel_tool_calls,
758 "presence_penalty": presence_penalty,
759 "response_format": response_format,
760 "seed": seed,
761 "service_tier": service_tier,
762 "stop": stop,
763 "store": store,
764 "stream": stream,
765 "stream_options": stream_options,
766 "temperature": temperature,
767 "tool_choice": tool_choice,
768 "tools": tools,
769 "top_logprobs": top_logprobs,
770 "top_p": top_p,
771 "user": user,
772 },
773 completion_create_params.CompletionCreateParams,
774 ),
775 options=make_request_options(
776 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
777 ),
778 cast_to=ChatCompletion,
779 stream=stream or False,
780 stream_cls=Stream[ChatCompletionChunk],
781 )
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_base_client.py:1270, in SyncAPIClient.post(self, path, cast_to, body, options, files, stream, stream_cls)
1256 def post(
1257 self,
1258 path: str,
(...)
1265 stream_cls: type[_StreamT] | None = None,
1266 ) -> ResponseT | _StreamT:
1267 opts = FinalRequestOptions.construct(
1268 method="post", url=path, json_data=body, files=to_httpx_files(files), **options
1269 )
-> 1270 return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_base_client.py:947, in SyncAPIClient.request(self, cast_to, options, remaining_retries, stream, stream_cls)
944 else:
945 retries_taken = 0
--> 947 return self._request(
948 cast_to=cast_to,
949 options=options,
950 stream=stream,
951 stream_cls=stream_cls,
952 retries_taken=retries_taken,
953 )
``````output
File ~/langchain/oss-py/docs/.venv/lib/python3.11/site-packages/openai/_base_client.py:1051, in SyncAPIClient._request(self, cast_to, options, retries_taken, stream, stream_cls)
1048 err.response.read()
1050 log.debug("Re-raising status error")
-> 1051 raise self._make_status_error_from_response(err.response) from None
1053 return self._process_response(
1054 cast_to=cast_to,
1055 options=options,
(...)
1059 retries_taken=retries_taken,
1060 )
``````output
BadRequestError: Error code: 400 - {'error': {'message': "Invalid parameter: messages with role 'tool' must be a response to a preceeding message with 'tool_calls'.", 'type': 'invalid_request_error', 'param': 'messages.[0].role', 'code': None}}
See this guide for more details on tool calling.
Troubleshooting
The following may help resolve this error:
- If you are using a custom executor rather than a prebuilt one like LangGraph's
ToolNodeor the legacy LangChain AgentExecutor, verify that you are invoking and returning the result for one tool per tool call. - If you are using few-shot tool call examples with messages that you manually create, and you want to simulate a failure,
you still need to pass back a
ToolMessagewhose content indicates that failure.