ollama._types.ResponseError ResponseError(e.response.text, e.response.status_code) from None

[Siki-cloud]

when i ran the following code， I got the error ollama._types.ResponseError， I don't why this happened, can anyone help me ? thanks in advance.

code

def test():
    resp = ollama.chat(
        model='llama3.1:8b', messages=[{'role': 'user', 'content': 'output the number from 1 to 5'}])  
    print(resp)

test()

here are the downloaded LLM：

/code/gencf# ollama list
NAME               ID              SIZE      MODIFIED
llama3.1:8b        46e0c10c039e    4.9 GB    53 minutes ago
llama3.1:latest    46e0c10c039e    4.9 GB    55 minutes ago
mistral:latest     f974a74358d6    4.1 GB    9 hours ago

Error：

Traceback (most recent call last):
  File "/code/gencf/multi_gpu_proc_generate.py", line 270, in <module>
    test()
  File "/code/gencf/multi_gpu_proc_generate.py", line 228, in test
    resp = ollama.chat(
  File "/root/anaconda3/envs/llmxq12/lib/python3.10/site-packages/ollama/_client.py", line 332, in chat
    return self._request(
  File "/root/anaconda3/envs/llmxq12/lib/python3.10/site-packages/ollama/_client.py", line 177, in _request
    return cls(**self._request_raw(*args, **kwargs).json())
  File "/root/anaconda3/envs/llmxq12/lib/python3.10/site-packages/ollama/_client.py", line 122, in _request_raw
    raise ResponseError(e.response.text, e.response.status_code) from None
ollama._types.ResponseError

[immmor]

Got the same error for Win10.

[Rakshitha7989]

getting the same error
raise ResponseError(e.response.text, e.response.status_code) from None
ollama._types.ResponseError: pull model manifest: file does not exist

[ParthSareen]

@Siki-cloud what's the text returned in the response error?

@Rakshitha7989 - are you able to run from CLI? ollama run <model>

[Zyg187]

I also encountered the same error in win10. I modified the location of the ollama model. Is it related to this reason?

**from ollama import chat
from ollama import ChatResponse

response: ChatResponse = chat(model='glm4:latest', messages=[
{
'role': 'user',
'content': 'Why is the sky blue?',
},
])
print(response['message']['content'])

or access fields directly from the response object

print(response.message.content)**

C:\Users\Administrator>ollama list
NAME ID SIZE MODIFIED
glm4:latest 5b699761eca5 5.5 GB 33 minutes ago
bge-m3:latest 790764642607 1.2 GB 4 weeks ago
EntropyYue/chatglm3:6b 8f6f34227356 3.6 GB 4 weeks ago

**D:\anaconda\envs\stu\python.exe D:\aaproject\learn-langchain\3.py
Traceback (most recent call last):
File "D:\aaproject\learn-langchain\3.py", line 4, in
response: ChatResponse = chat(model='glm4:latest', messages=[
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\anaconda\envs\stu\Lib\site-packages\ollama_client.py", line 332, in chat
return self._request(
^^^^^^^^^^^^^^
File "D:\anaconda\envs\stu\Lib\site-packages\ollama_client.py", line 177, in _request
return cls(**self._request_raw(*args, kwargs).json())
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\anaconda\envs\stu\Lib\site-packages\ollama_client.py", line 122, in _request_raw
raise ResponseError(e.response.text, e.response.status_code) from None
ollama._types.ResponseError

[Zyg187]

But I can load the model using from langchain_community.llms import Ollama. This method does not report an error, why is this?

`# 使用本地ollama调用模型，简单实现一个链式回答
from langchain_core.prompts import ChatPromptTemplate
from langchain_community.llms import Ollama
from langchain_core.output_parsers import StrOutputParser
model = Ollama(model="glm4:latest", temperature=0.7) # 调用模型
output_parser = StrOutputParser() # 解析格式

模板

template = """
你是一个乐于助人的AI，擅长于解决回答各种问题。
问题：{question}
"""
prompt = ChatPromptTemplate.from_template(template)
chain = prompt | model | output_parser
print(chain.invoke({"question": "你比GPT4厉害吗？"}))`

[ParthSareen]

Rolling out a change with some better error messaging. Until then could all of you make sure that ollama run <model> <prompt> is working?

[Zyg187]

Yes, I was able to use ollama in cmd, and I was able to use ollama via langchain's from langchain_community.llms import Ollama, but I get an error when I use the ollama-python library. I will provide more information but need guidance on what to do

[ParthSareen]

@Zyg187 could you try passing in the url when instantiating the ollama client under host?

[Zyg187]

@ParthSareen Sorry to keep you waiting. I'm a novice who is new to contact, I don't quite understand what you mean, can you provide a simple piece of code for me to test?

[Zyg187]

@ParthSareen I understand what you mean, and I have found the cause of the problem. Because I opened a proxy, I can't access it locally.The code runs successfully after turning off the proxy. I am curious about why the method "from langchain_community.llms import Ollama" can be used normally when using a proxy. How can I use the ollama-python official library normally when opening a proxy? Can you give me some suggestions?

from ollama import Client client = Client( host='http://localhost:11434', headers={'x-some-header': 'some-value'} ) response = client.chat(model='llama3.2', messages=[ { 'role': 'user', 'content': 'Why is the sky blue?', }, ])

[ParthSareen]

Yeah unsure why langchain works - I wonder if it's not the proxied instance. Could you check this out and see if it helps? https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-use-ollama-behind-a-proxy

[ParthSareen]

Going to close for now, feel free to ping if you're still running into this!